gzdoom-gles/src/thingdef/thingdef_properties.cpp

2933 lines
99 KiB
C++
Raw Normal View History

/*
** thingdef-properties.cpp
**
** Actor definitions - properties and flags parser
**
**---------------------------------------------------------------------------
** Copyright 2002-2007 Christoph Oelckers
** Copyright 2004-2007 Randy Heit
** All rights reserved.
**
** Redistribution and use in source and binary forms, with or without
** modification, are permitted provided that the following conditions
** are met:
**
** 1. Redistributions of source code must retain the above copyright
** notice, this list of conditions and the following disclaimer.
** 2. Redistributions in binary form must reproduce the above copyright
** notice, this list of conditions and the following disclaimer in the
** documentation and/or other materials provided with the distribution.
** 3. The name of the author may not be used to endorse or promote products
** derived from this software without specific prior written permission.
** 4. When not used as part of ZDoom or a ZDoom derivative, this code will be
** covered by the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or (at
** your option) any later version.
**
** THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
** IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
** OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
** IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
** INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
** NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
** DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
** (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
** THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
**---------------------------------------------------------------------------
**
*/
#include "gi.h"
#include "actor.h"
#include "info.h"
#include "sc_man.h"
#include "tarray.h"
#include "w_wad.h"
#include "templates.h"
#include "r_defs.h"
#include "r_draw.h"
#include "a_pickups.h"
#include "s_sound.h"
#include "cmdlib.h"
#include "p_lnspec.h"
#include "a_action.h"
#include "decallib.h"
#include "m_random.h"
#include "i_system.h"
#include "p_local.h"
#include "p_effect.h"
#include "v_palette.h"
#include "doomerrors.h"
#include "a_hexenglobal.h"
#include "a_weaponpiece.h"
#include "p_conversation.h"
#include "v_text.h"
#include "thingdef.h"
#include "a_sharedglobal.h"
#include "r_translate.h"
#include "a_morph.h"
#include "colormatcher.h"
//==========================================================================
//
// List of all flags
//
//==========================================================================
// [RH] Keep GCC quiet by not using offsetof on Actor types.
#define DEFINE_FLAG(prefix, name, type, variable) { prefix##_##name, #name, (int)(size_t)&((type*)1)->variable - 1 }
#define DEFINE_FLAG2(symbol, name, type, variable) { symbol, #name, (int)(size_t)&((type*)1)->variable - 1 }
#define DEFINE_DEPRECATED_FLAG(name) { DEPF_##name, #name, -1 }
#define DEFINE_DUMMY_FLAG(name) { DEPF_UNUSED, #name, -1 }
struct flagdef
{
int flagbit;
const char *name;
int structoffset;
};
enum
{
DEPF_UNUSED,
DEPF_FIREDAMAGE,
DEPF_ICEDAMAGE,
DEPF_LOWGRAVITY,
DEPF_LONGMELEERANGE,
DEPF_SHORTMISSILERANGE,
DEPF_PICKUPFLASH,
DEPF_QUARTERGRAVITY,
DEPF_FIRERESIST,
};
static flagdef ActorFlags[]=
{
DEFINE_FLAG(MF, PICKUP, APlayerPawn, flags),
DEFINE_FLAG(MF, SPECIAL, APlayerPawn, flags),
DEFINE_FLAG(MF, SOLID, AActor, flags),
DEFINE_FLAG(MF, SHOOTABLE, AActor, flags),
DEFINE_FLAG(MF, NOSECTOR, AActor, flags),
DEFINE_FLAG(MF, NOBLOCKMAP, AActor, flags),
DEFINE_FLAG(MF, AMBUSH, AActor, flags),
DEFINE_FLAG(MF, JUSTHIT, AActor, flags),
DEFINE_FLAG(MF, JUSTATTACKED, AActor, flags),
DEFINE_FLAG(MF, SPAWNCEILING, AActor, flags),
DEFINE_FLAG(MF, NOGRAVITY, AActor, flags),
DEFINE_FLAG(MF, DROPOFF, AActor, flags),
DEFINE_FLAG(MF, NOCLIP, AActor, flags),
DEFINE_FLAG(MF, FLOAT, AActor, flags),
DEFINE_FLAG(MF, TELEPORT, AActor, flags),
DEFINE_FLAG(MF, MISSILE, AActor, flags),
DEFINE_FLAG(MF, DROPPED, AActor, flags),
DEFINE_FLAG(MF, SHADOW, AActor, flags),
DEFINE_FLAG(MF, NOBLOOD, AActor, flags),
DEFINE_FLAG(MF, CORPSE, AActor, flags),
DEFINE_FLAG(MF, INFLOAT, AActor, flags),
DEFINE_FLAG(MF, COUNTKILL, AActor, flags),
DEFINE_FLAG(MF, COUNTITEM, AActor, flags),
DEFINE_FLAG(MF, SKULLFLY, AActor, flags),
DEFINE_FLAG(MF, NOTDMATCH, AActor, flags),
DEFINE_FLAG(MF, SPAWNSOUNDSOURCE, AActor, flags),
DEFINE_FLAG(MF, FRIENDLY, AActor, flags),
DEFINE_FLAG(MF, NOLIFTDROP, AActor, flags),
DEFINE_FLAG(MF, STEALTH, AActor, flags),
DEFINE_FLAG(MF, ICECORPSE, AActor, flags),
DEFINE_FLAG(MF2, DONTREFLECT, AActor, flags2),
DEFINE_FLAG(MF2, WINDTHRUST, AActor, flags2),
DEFINE_FLAG(MF2, HERETICBOUNCE , AActor, flags2),
DEFINE_FLAG(MF2, BLASTED, AActor, flags2),
DEFINE_FLAG(MF2, FLOORCLIP, AActor, flags2),
DEFINE_FLAG(MF2, SPAWNFLOAT, AActor, flags2),
DEFINE_FLAG(MF2, NOTELEPORT, AActor, flags2),
DEFINE_FLAG2(MF2_RIP, RIPPER, AActor, flags2),
DEFINE_FLAG(MF2, PUSHABLE, AActor, flags2),
DEFINE_FLAG2(MF2_SLIDE, SLIDESONWALLS, AActor, flags2),
DEFINE_FLAG2(MF2_PASSMOBJ, CANPASS, AActor, flags2),
DEFINE_FLAG(MF2, CANNOTPUSH, AActor, flags2),
DEFINE_FLAG(MF2, THRUGHOST, AActor, flags2),
DEFINE_FLAG(MF2, BOSS, AActor, flags2),
DEFINE_FLAG2(MF2_NODMGTHRUST, NODAMAGETHRUST, AActor, flags2),
DEFINE_FLAG(MF2, DONTTRANSLATE, AActor, flags2),
DEFINE_FLAG(MF2, TELESTOMP, AActor, flags2),
DEFINE_FLAG(MF2, FLOATBOB, AActor, flags2),
DEFINE_FLAG(MF2, HEXENBOUNCE, AActor, flags2),
DEFINE_FLAG(MF2, DOOMBOUNCE, AActor, flags2),
DEFINE_FLAG2(MF2_IMPACT, ACTIVATEIMPACT, AActor, flags2),
DEFINE_FLAG2(MF2_PUSHWALL, CANPUSHWALLS, AActor, flags2),
DEFINE_FLAG2(MF2_MCROSS, ACTIVATEMCROSS, AActor, flags2),
DEFINE_FLAG2(MF2_PCROSS, ACTIVATEPCROSS, AActor, flags2),
DEFINE_FLAG(MF2, CANTLEAVEFLOORPIC, AActor, flags2),
DEFINE_FLAG(MF2, NONSHOOTABLE, AActor, flags2),
DEFINE_FLAG(MF2, INVULNERABLE, AActor, flags2),
DEFINE_FLAG(MF2, DORMANT, AActor, flags2),
DEFINE_FLAG(MF2, SEEKERMISSILE, AActor, flags2),
DEFINE_FLAG(MF2, REFLECTIVE, AActor, flags2),
DEFINE_FLAG(MF3, FLOORHUGGER, AActor, flags3),
DEFINE_FLAG(MF3, CEILINGHUGGER, AActor, flags3),
DEFINE_FLAG(MF3, NORADIUSDMG, AActor, flags3),
DEFINE_FLAG(MF3, GHOST, AActor, flags3),
DEFINE_FLAG(MF3, SPECIALFLOORCLIP, AActor, flags3),
DEFINE_FLAG(MF3, ALWAYSPUFF, AActor, flags3),
DEFINE_FLAG(MF3, DONTSPLASH, AActor, flags3),
DEFINE_FLAG(MF3, DONTOVERLAP, AActor, flags3),
DEFINE_FLAG(MF3, DONTMORPH, AActor, flags3),
DEFINE_FLAG(MF3, DONTSQUASH, AActor, flags3),
DEFINE_FLAG(MF3, EXPLOCOUNT, AActor, flags3),
DEFINE_FLAG(MF3, FULLVOLACTIVE, AActor, flags3),
DEFINE_FLAG(MF3, ISMONSTER, AActor, flags3),
DEFINE_FLAG(MF3, SKYEXPLODE, AActor, flags3),
DEFINE_FLAG(MF3, STAYMORPHED, AActor, flags3),
DEFINE_FLAG(MF3, DONTBLAST, AActor, flags3),
DEFINE_FLAG(MF3, CANBLAST, AActor, flags3),
DEFINE_FLAG(MF3, NOTARGET, AActor, flags3),
DEFINE_FLAG(MF3, DONTGIB, AActor, flags3),
DEFINE_FLAG(MF3, NOBLOCKMONST, AActor, flags3),
DEFINE_FLAG(MF3, FULLVOLDEATH, AActor, flags3),
DEFINE_FLAG(MF3, CANBOUNCEWATER, AActor, flags3),
DEFINE_FLAG(MF3, NOWALLBOUNCESND, AActor, flags3),
DEFINE_FLAG(MF3, FOILINVUL, AActor, flags3),
DEFINE_FLAG(MF3, NOTELEOTHER, AActor, flags3),
DEFINE_FLAG(MF3, BLOODLESSIMPACT, AActor, flags3),
DEFINE_FLAG(MF3, NOEXPLODEFLOOR, AActor, flags3),
DEFINE_FLAG(MF3, PUFFONACTORS, AActor, flags3),
DEFINE_FLAG(MF4, QUICKTORETALIATE, AActor, flags4),
DEFINE_FLAG(MF4, NOICEDEATH, AActor, flags4),
DEFINE_FLAG(MF4, RANDOMIZE, AActor, flags4),
DEFINE_FLAG(MF4, FIXMAPTHINGPOS , AActor, flags4),
DEFINE_FLAG(MF4, ACTLIKEBRIDGE, AActor, flags4),
DEFINE_FLAG(MF4, STRIFEDAMAGE, AActor, flags4),
DEFINE_FLAG(MF4, CANUSEWALLS, AActor, flags4),
DEFINE_FLAG(MF4, MISSILEMORE, AActor, flags4),
DEFINE_FLAG(MF4, MISSILEEVENMORE, AActor, flags4),
DEFINE_FLAG(MF4, FORCERADIUSDMG, AActor, flags4),
DEFINE_FLAG(MF4, DONTFALL, AActor, flags4),
DEFINE_FLAG(MF4, SEESDAGGERS, AActor, flags4),
DEFINE_FLAG(MF4, INCOMBAT, AActor, flags4),
DEFINE_FLAG(MF4, LOOKALLAROUND, AActor, flags4),
DEFINE_FLAG(MF4, STANDSTILL, AActor, flags4),
DEFINE_FLAG(MF4, SPECTRAL, AActor, flags4),
DEFINE_FLAG(MF4, NOSPLASHALERT, AActor, flags4),
DEFINE_FLAG(MF4, SYNCHRONIZED, AActor, flags4),
DEFINE_FLAG(MF4, NOTARGETSWITCH, AActor, flags4),
DEFINE_FLAG(MF4, DONTHURTSPECIES, AActor, flags4),
DEFINE_FLAG(MF4, SHIELDREFLECT, AActor, flags4),
DEFINE_FLAG(MF4, DEFLECT, AActor, flags4),
DEFINE_FLAG(MF4, ALLOWPARTICLES, AActor, flags4),
DEFINE_FLAG(MF4, EXTREMEDEATH, AActor, flags4),
DEFINE_FLAG(MF4, NOEXTREMEDEATH, AActor, flags4),
DEFINE_FLAG(MF4, FRIGHTENED, AActor, flags4),
DEFINE_FLAG(MF4, NOBOUNCESOUND, AActor, flags4),
DEFINE_FLAG(MF4, NOSKIN, AActor, flags4),
DEFINE_FLAG(MF4, BOSSDEATH, AActor, flags4),
DEFINE_FLAG(MF5, FASTER, AActor, flags5),
DEFINE_FLAG(MF5, FASTMELEE, AActor, flags5),
DEFINE_FLAG(MF5, NODROPOFF, AActor, flags5),
DEFINE_FLAG(MF5, BOUNCEONACTORS, AActor, flags5),
DEFINE_FLAG(MF5, EXPLODEONWATER, AActor, flags5),
DEFINE_FLAG(MF5, NODAMAGE, AActor, flags5),
DEFINE_FLAG(MF5, BLOODSPLATTER, AActor, flags5),
DEFINE_FLAG(MF5, OLDRADIUSDMG, AActor, flags5),
DEFINE_FLAG(MF5, DEHEXPLOSION, AActor, flags5),
DEFINE_FLAG(MF5, PIERCEARMOR, AActor, flags5),
DEFINE_FLAG(MF5, NOBLOODDECALS, AActor, flags5),
DEFINE_FLAG(MF5, USESPECIAL, AActor, flags5),
DEFINE_FLAG(MF5, NOPAIN, AActor, flags5),
DEFINE_FLAG(MF5, ALWAYSFAST, AActor, flags5),
DEFINE_FLAG(MF5, NEVERFAST, AActor, flags5),
DEFINE_FLAG(MF5, ALWAYSRESPAWN, AActor, flags5),
DEFINE_FLAG(MF5, NEVERRESPAWN, AActor, flags5),
DEFINE_FLAG(MF5, DONTRIP, AActor, flags5),
DEFINE_FLAG(MF5, NOINFIGHTING, AActor, flags5),
DEFINE_FLAG(MF5, NOINTERACTION, AActor, flags5),
DEFINE_FLAG(MF5, NOTIMEFREEZE, AActor, flags5),
DEFINE_FLAG(MF5, PUFFGETSOWNER, AActor, flags5), // [BB] added PUFFGETSOWNER
DEFINE_FLAG(MF5, SPECIALFIREDAMAGE, AActor, flags5),
DEFINE_FLAG(MF5, SUMMONEDMONSTER, AActor, flags5),
DEFINE_FLAG(MF5, NOVERTICALMELEERANGE, AActor, flags5),
// Effect flags
DEFINE_FLAG(FX, VISIBILITYPULSE, AActor, effects),
DEFINE_FLAG2(FX_ROCKET, ROCKETTRAIL, AActor, effects),
DEFINE_FLAG2(FX_GRENADE, GRENADETRAIL, AActor, effects),
DEFINE_FLAG(RF, INVISIBLE, AActor, renderflags),
DEFINE_FLAG(RF, FORCEYBILLBOARD, AActor, renderflags),
DEFINE_FLAG(RF, FORCEXYBILLBOARD, AActor, renderflags),
// Deprecated flags. Handling must be performed in HandleDeprecatedFlags
DEFINE_DEPRECATED_FLAG(FIREDAMAGE),
DEFINE_DEPRECATED_FLAG(ICEDAMAGE),
DEFINE_DEPRECATED_FLAG(LOWGRAVITY),
DEFINE_DEPRECATED_FLAG(SHORTMISSILERANGE),
DEFINE_DEPRECATED_FLAG(LONGMELEERANGE),
DEFINE_DEPRECATED_FLAG(QUARTERGRAVITY),
DEFINE_DEPRECATED_FLAG(FIRERESIST),
DEFINE_DUMMY_FLAG(NONETID),
DEFINE_DUMMY_FLAG(ALLOWCLIENTSPAWN),
};
static flagdef InventoryFlags[] =
{
// Inventory flags
DEFINE_FLAG(IF, QUIET, AInventory, ItemFlags),
DEFINE_FLAG(IF, AUTOACTIVATE, AInventory, ItemFlags),
DEFINE_FLAG(IF, UNDROPPABLE, AInventory, ItemFlags),
DEFINE_FLAG(IF, INVBAR, AInventory, ItemFlags),
DEFINE_FLAG(IF, HUBPOWER, AInventory, ItemFlags),
DEFINE_FLAG(IF, INTERHUBSTRIP, AInventory, ItemFlags),
DEFINE_FLAG(IF, ALWAYSPICKUP, AInventory, ItemFlags),
DEFINE_FLAG(IF, FANCYPICKUPSOUND, AInventory, ItemFlags),
DEFINE_FLAG(IF, BIGPOWERUP, AInventory, ItemFlags),
DEFINE_FLAG(IF, KEEPDEPLETED, AInventory, ItemFlags),
DEFINE_FLAG(IF, IGNORESKILL, AInventory, ItemFlags),
DEFINE_FLAG(IF, ADDITIVETIME, AInventory, ItemFlags),
DEFINE_FLAG(IF, NOATTENPICKUPSOUND, AInventory, ItemFlags),
DEFINE_DEPRECATED_FLAG(PICKUPFLASH),
};
static flagdef WeaponFlags[] =
{
// Weapon flags
DEFINE_FLAG(WIF, NOAUTOFIRE, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, READYSNDHALF, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, DONTBOB, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, AXEBLOOD, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, NOALERT, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, AMMO_OPTIONAL, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, ALT_AMMO_OPTIONAL, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, PRIMARY_USES_BOTH, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, WIMPY_WEAPON, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, POWERED_UP, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, STAFF2_KICKBACK, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF_BOT, EXPLOSIVE, AWeapon, WeaponFlags),
DEFINE_FLAG2(WIF_BOT_MELEE, MELEEWEAPON, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF_BOT, BFG, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, CHEATNOTWEAPON, AWeapon, WeaponFlags),
DEFINE_FLAG(WIF, NO_AUTO_SWITCH, AWeapon, WeaponFlags),
DEFINE_DUMMY_FLAG(NOLMS),
};
static const struct { const PClass *Type; flagdef *Defs; int NumDefs; } FlagLists[] =
{
{ RUNTIME_CLASS(AActor), ActorFlags, sizeof(ActorFlags)/sizeof(flagdef) },
{ RUNTIME_CLASS(AInventory), InventoryFlags, sizeof(InventoryFlags)/sizeof(flagdef) },
{ RUNTIME_CLASS(AWeapon), WeaponFlags, sizeof(WeaponFlags)/sizeof(flagdef) }
};
#define NUM_FLAG_LISTS 3
//==========================================================================
//
// Find a flag by name using a binary search
//
//==========================================================================
static int STACK_ARGS flagcmp (const void * a, const void * b)
{
return stricmp( ((flagdef*)a)->name, ((flagdef*)b)->name);
}
static flagdef *FindFlag (flagdef *flags, int numflags, const char *flag)
{
int min = 0, max = numflags - 1;
while (min <= max)
{
int mid = (min + max) / 2;
int lexval = stricmp (flag, flags[mid].name);
if (lexval == 0)
{
return &flags[mid];
}
else if (lexval > 0)
{
min = mid + 1;
}
else
{
max = mid - 1;
}
}
return NULL;
}
static flagdef *FindFlag (const PClass *type, const char *part1, const char *part2)
{
static bool flagsorted = false;
flagdef *def;
int i;
if (!flagsorted)
{
for (i = 0; i < NUM_FLAG_LISTS; ++i)
{
qsort (FlagLists[i].Defs, FlagLists[i].NumDefs, sizeof(flagdef), flagcmp);
}
flagsorted = true;
}
if (part2 == NULL)
{ // Search all lists
for (i = 0; i < NUM_FLAG_LISTS; ++i)
{
if (type->IsDescendantOf (FlagLists[i].Type))
{
def = FindFlag (FlagLists[i].Defs, FlagLists[i].NumDefs, part1);
if (def != NULL)
{
return def;
}
}
}
}
else
{ // Search just the named list
for (i = 0; i < NUM_FLAG_LISTS; ++i)
{
if (stricmp (FlagLists[i].Type->TypeName.GetChars(), part1) == 0)
{
if (type->IsDescendantOf (FlagLists[i].Type))
{
return FindFlag (FlagLists[i].Defs, FlagLists[i].NumDefs, part2);
}
else
{
return NULL;
}
}
}
}
return NULL;
}
//===========================================================================
//
// HandleDeprecatedFlags
//
// Handles the deprecated flags and sets the respective properties
// to appropriate values. This is solely intended for backwards
// compatibility so mixing this with code that is aware of the real
// properties is not recommended
//
//===========================================================================
static void HandleDeprecatedFlags(AActor *defaults, FActorInfo *info, bool set, int index)
{
switch (index)
{
case DEPF_FIREDAMAGE:
defaults->DamageType = set? NAME_Fire : NAME_None;
break;
case DEPF_ICEDAMAGE:
defaults->DamageType = set? NAME_Ice : NAME_None;
break;
case DEPF_LOWGRAVITY:
defaults->gravity = set? FRACUNIT/8 : FRACUNIT;
break;
case DEPF_SHORTMISSILERANGE:
defaults->maxtargetrange = set? 896*FRACUNIT : 0;
break;
case DEPF_LONGMELEERANGE:
defaults->meleethreshold = set? 196*FRACUNIT : 0;
break;
case DEPF_QUARTERGRAVITY:
defaults->gravity = set? FRACUNIT/4 : FRACUNIT;
break;
case DEPF_FIRERESIST:
info->SetDamageFactor(NAME_Fire, set? FRACUNIT/2 : FRACUNIT);
break;
case DEPF_PICKUPFLASH:
if (set)
{
static_cast<AInventory*>(defaults)->PickupFlash = fuglyname("PickupFlash");
}
else
{
static_cast<AInventory*>(defaults)->PickupFlash = NULL;
}
break;
default:
break; // silence GCC
}
}
//===========================================================================
//
// A_ChangeFlag
//
// This cannot be placed in thingdef_codeptr because it needs the flag table
//
//===========================================================================
DEFINE_ACTION_FUNCTION_PARAMS(AActor, A_ChangeFlag)
{
ACTION_PARAM_START(2);
ACTION_PARAM_STRING(flagname, 0);
ACTION_PARAM_BOOL(expression, 1);
const char *dot = strchr (flagname, '.');
flagdef *fd;
const PClass *cls = self->GetClass();
if (dot != NULL)
{
FString part1(flagname, dot-flagname);
fd = FindFlag (cls, part1, dot+1);
}
else
{
fd = FindFlag (cls, flagname, NULL);
}
if (fd != NULL)
{
if (fd->structoffset == -1)
{
HandleDeprecatedFlags(self, cls->ActorInfo, expression, fd->flagbit);
}
else
{
int * flagp = (int*) (((char*)self) + fd->structoffset);
if (expression) *flagp |= fd->flagbit;
else *flagp &= ~fd->flagbit;
}
}
else
{
Printf("Unknown flag '%s' in '%s'\n", flagname, cls->TypeName.GetChars());
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
void ParseActorFlag (FScanner &sc, Baggage &bag, int mod)
{
flagdef *fd;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
FString part1 = sc.String;
const char *part2 = NULL;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString ("."))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
part2 = sc.String;
}
if ( (fd = FindFlag (bag.Info->Class, part1.GetChars(), part2)) )
{
AActor *defaults = (AActor*)bag.Info->Class->Defaults;
if (fd->structoffset == -1) // this is a deprecated flag that has been changed into a real property
{
HandleDeprecatedFlags(defaults, bag.Info, mod=='+', fd->flagbit);
}
else
{
DWORD * flagvar = (DWORD*) ((char*)defaults + fd->structoffset);
if (mod == '+')
{
*flagvar |= fd->flagbit;
}
else
{
*flagvar &= ~fd->flagbit;
}
}
}
else
{
if (part2 == NULL)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("\"%s\" is an unknown flag\n", part1.GetChars());
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("\"%s.%s\" is an unknown flag\n", part1.GetChars(), part2);
}
}
}
//----------------------------------------------------------------------------
//----------------------------------------------------------------------------
//
// Translation parsing
//
//----------------------------------------------------------------------------
//----------------------------------------------------------------------------
FRemapTable CurrentTranslation;
TArray<PalEntry> BloodTranslationColors;
PalEntry BloodTranslations[256];
static bool Check(char *& range, char c, bool error=true)
{
while (isspace(*range)) range++;
if (*range==c)
{
range++;
return true;
}
if (error)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
//sc.ScriptError("Invalid syntax in translation specification: '%c' expected", c);
}
return false;
}
static void AddToTranslation(char * range)
{
int start,end;
start=strtol(range, &range, 10);
if (!Check(range, ':')) return;
end=strtol(range, &range, 10);
if (!Check(range, '=')) return;
if (!Check(range, '[', false))
{
int pal1,pal2;
pal1=strtol(range, &range, 10);
if (!Check(range, ':')) return;
pal2=strtol(range, &range, 10);
CurrentTranslation.AddIndexRange(start, end, pal1, pal2);
}
else
{
// translation using RGB values
int r1,g1,b1,r2,g2,b2;
r1=strtol(range, &range, 10);
if (!Check(range, ',')) return;
g1=strtol(range, &range, 10);
if (!Check(range, ',')) return;
b1=strtol(range, &range, 10);
if (!Check(range, ']')) return;
if (!Check(range, ':')) return;
if (!Check(range, '[')) return;
r2=strtol(range, &range, 10);
if (!Check(range, ',')) return;
g2=strtol(range, &range, 10);
if (!Check(range, ',')) return;
b2=strtol(range, &range, 10);
if (!Check(range, ']')) return;
CurrentTranslation.AddColorRange(start, end, r1, g1, b1, r2, g2, b2);
}
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static int StoreTranslation(FScanner &sc)
{
unsigned int i;
for (i = 0; i < translationtables[TRANSLATION_Decorate].Size(); i++)
{
if (CurrentTranslation == *translationtables[TRANSLATION_Decorate][i])
{
// A duplicate of this translation already exists
return TRANSLATION(TRANSLATION_Decorate, i);
}
}
if (translationtables[TRANSLATION_Decorate].Size() >= MAX_DECORATE_TRANSLATIONS)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("Too many translations in DECORATE");
}
FRemapTable *newtrans = new FRemapTable;
*newtrans = CurrentTranslation;
i = translationtables[TRANSLATION_Decorate].Push(newtrans);
return TRANSLATION(TRANSLATION_Decorate, i);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static int CreateBloodTranslation(FScanner &sc, PalEntry color)
{
unsigned int i;
for (i = 0; i < BloodTranslationColors.Size(); i++)
{
if (color.r == BloodTranslationColors[i].r &&
color.g == BloodTranslationColors[i].g &&
color.b == BloodTranslationColors[i].b)
{
// A duplicate of this translation already exists
return i;
}
}
if (BloodTranslationColors.Size() >= MAX_DECORATE_TRANSLATIONS)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("Too many blood colors in DECORATE");
}
FRemapTable *trans = new FRemapTable;
for (i = 0; i < 256; i++)
{
int bright = MAX(MAX(GPalette.BaseColors[i].r, GPalette.BaseColors[i].g), GPalette.BaseColors[i].b);
PalEntry pe = PalEntry(color.r*bright/255, color.g*bright/255, color.b*bright/255);
int entry = ColorMatcher.Pick(pe.r, pe.g, pe.b);
trans->Palette[i] = pe;
trans->Remap[i] = entry;
}
translationtables[TRANSLATION_Blood].Push(trans);
return BloodTranslationColors.Push(color);
}
//----------------------------------------------------------------------------
//
// DropItem handling
//
//----------------------------------------------------------------------------
static void FreeDropItemChain(FDropItem *chain)
{
while (chain != NULL)
{
FDropItem *next = chain->Next;
delete chain;
chain = next;
}
}
class FDropItemPtrArray : public TArray<FDropItem *>
{
public:
~FDropItemPtrArray()
{
for (unsigned int i = 0; i < Size(); ++i)
{
FreeDropItemChain ((*this)[i]);
}
}
};
static FDropItemPtrArray DropItemList;
FDropItem *GetDropItems(const PClass *cls)
{
unsigned int index = cls->Meta.GetMetaInt (ACMETA_DropItems) - 1;
if (index >= 0 && index < DropItemList.Size())
{
return DropItemList[index];
}
return NULL;
}
//==========================================================================
//
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
typedef void (*ActorPropFunction) (FScanner &sc, AActor *defaults, Baggage &bag);
struct ActorProps
{
const char *name;
ActorPropFunction Handler;
const PClass * type;
};
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
typedef ActorProps (*ActorPropHandler) (const char *str, unsigned int len);
static const ActorProps *is_actorprop (const char *str);
//==========================================================================
//
// Checks for a numeric parameter which may or may not be preceded by a comma
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static bool CheckNumParm(FScanner &sc)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString(","))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
return true;
}
else
{
return sc.CheckNumber();
}
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static bool CheckFloatParm(FScanner &sc)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString(","))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
return true;
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
return sc.CheckFloat();
}
}
// [MH]
static int ParseMorphStyle (FScanner &sc)
{
static const char * morphstyles[]={
"MRF_ADDSTAMINA", "MRF_FULLHEALTH", "MRF_UNDOBYTOMEOFPOWER", "MRF_UNDOBYCHAOSDEVICE",
"MRF_FAILNOTELEFRAG", "MRF_FAILNOLAUGH", "MRF_WHENINVULNERABLE", "MRF_LOSEACTUALWEAPON",
"MRF_NEWTIDBEHAVIOUR", "MRF_UNDOBYDEATH", "MRF_UNDOBYDEATHFORCED", "MRF_UNDOBYDEATHSAVES", NULL};
static const int morphstyle_values[]={
MORPH_ADDSTAMINA, MORPH_FULLHEALTH, MORPH_UNDOBYTOMEOFPOWER, MORPH_UNDOBYCHAOSDEVICE,
MORPH_FAILNOTELEFRAG, MORPH_FAILNOLAUGH, MORPH_WHENINVULNERABLE, MORPH_LOSEACTUALWEAPON,
MORPH_NEWTIDBEHAVIOUR, MORPH_UNDOBYDEATH, MORPH_UNDOBYDEATHFORCED, MORPH_UNDOBYDEATHSAVES};
// May be given flags by number...
if (sc.CheckNumber())
{
sc.MustGetNumber();
return sc.Number;
}
// ... else should be flags by name.
// NOTE: Later this should be removed and a normal expression used.
// The current DECORATE parser can't handle this though.
bool gotparen = sc.CheckString("(");
int style = 0;
do
{
sc.MustGetString();
style |= morphstyle_values[sc.MustMatchString(morphstyles)];
}
while (sc.CheckString("|"));
if (gotparen)
{
sc.MustGetStringName(")");
}
return style;
}
//==========================================================================
//
// Property parsers
//
//==========================================================================
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorSkipSuper (FScanner &sc, AActor *defaults, Baggage &bag)
{
if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(AInventory)))
{
sc.ScriptMessage("'skip_super' in definition of inventory item igmored.\n");
return;
}
memcpy (defaults, GetDefault<AActor>(), sizeof(AActor));
if (bag.DropItemList != NULL)
{
FreeDropItemChain (bag.DropItemList);
}
ResetBaggage (&bag);
MakeStateDefines(NULL);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorGame (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
if (sc.Compare ("Doom"))
{
bag.Info->GameFilter |= GAME_Doom;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Heretic"))
{
bag.Info->GameFilter |= GAME_Heretic;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Hexen"))
{
bag.Info->GameFilter |= GAME_Hexen;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Raven"))
{
bag.Info->GameFilter |= GAME_Raven;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Strife"))
{
bag.Info->GameFilter |= GAME_Strife;
}
else if (sc.Compare ("Chex"))
{
bag.Info->GameFilter |= GAME_Chex;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Any"))
{
bag.Info->GameFilter = GAME_Any;
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("Unknown game type %s", sc.String);
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorSpawnID (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
if (sc.Number<0 || sc.Number>255)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("SpawnID must be in the range [0,255]");
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else bag.Info->SpawnID=(BYTE)sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorConversationID (FScanner &sc, AActor *defaults, Baggage &bag)
{
int convid;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
convid = sc.Number;
// Handling for Strife teaser IDs - only of meaning for the standard items
// as PWADs cannot be loaded with the teasers.
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString(","))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
if ((gameinfo.flags & (GI_SHAREWARE|GI_TEASER2)) == (GI_SHAREWARE))
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
convid=sc.Number;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetStringName(",");
sc.MustGetNumber();
if ((gameinfo.flags & (GI_SHAREWARE|GI_TEASER2)) == (GI_SHAREWARE|GI_TEASER2))
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
convid=sc.Number;
if (convid==-1) return;
}
if (convid<0 || convid>1000)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("ConversationID must be in the range [0,1000]");
}
else StrifeTypes[convid] = bag.Info->Class;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorTag (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaString(AMETA_StrifeName, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorHealth (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->health=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorGibHealth (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (AMETA_GibHealth, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorWoundHealth (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (AMETA_WoundHealth, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorReactionTime (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->reactiontime=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorPainChance (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (!sc.CheckNumber())
{
FName painType;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
if (sc.Compare("Normal")) painType = NAME_None;
else painType=sc.String;
sc.MustGetToken(',');
sc.MustGetNumber();
bag.Info->SetPainChance(painType, sc.Number);
return;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->PainChance=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDamage (FScanner &sc, AActor *defaults, Baggage &bag)
{
// Damage can either be a single number, in which case it is subject
// to the original damage calculation rules. Or, it can be an expression
// and will be calculated as-is, ignoring the original rules. For
// compatibility reasons, expressions must be enclosed within
// parentheses.
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString ("("))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->Damage = 0x40000000 | ParseExpression (sc, false, bag.Info->Class);
sc.MustGetStringName(")");
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->Damage = sc.Number;
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorSpeed (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->Speed=fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorFloatSpeed (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->FloatSpeed=fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorRadius (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->radius=fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorHeight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->height=fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMass (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->Mass=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorXScale (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->scaleX = FLOAT2FIXED(sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorYScale (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->scaleY = FLOAT2FIXED(sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorScale (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->scaleX= defaults->scaleY = FLOAT2FIXED(sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorArgs (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
for (int i = 0; i < 5; i++)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->args[i] = sc.Number;
if (i < 4 && !sc.CheckToken(',')) break;
}
defaults->flags2|=MF2_ARGSDEFINED;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorSeeSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->SeeSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorAttackSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->AttackSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorPainSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->PainSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDeathSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->DeathSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorActiveSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->ActiveSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorHowlSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaInt (AMETA_HowlSound, S_FindSound(sc.String));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDropItem (FScanner &sc, AActor *defaults, Baggage &bag)
{
// create a linked list of dropitems
if (!bag.DropItemSet)
{
bag.DropItemSet = true;
bag.DropItemList = NULL;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
FDropItem *di = new FDropItem;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
di->Name=sc.String;
di->probability=255;
di->amount=-1;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (CheckNumParm(sc))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
di->probability = sc.Number;
if (CheckNumParm(sc))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
di->amount = sc.Number;
}
}
di->Next = bag.DropItemList;
bag.DropItemList = di;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorSpawnState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Spawn", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorSeeState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("See", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMeleeState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Melee", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMissileState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Missile", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorPainState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Pain", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDeathState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Death", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorXDeathState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("XDeath", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorBurnState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Burn", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorIceState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Ice", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorRaiseState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Raise", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorCrashState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Crash", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorCrushState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Crush", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorWoundState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Wound", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDisintegrateState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Disintegrate", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorHealState (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
AddState("Heal", CheckState (sc, bag.Info->Class));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorStates (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (!bag.StateSet) ParseStates(sc, bag.Info, defaults, bag);
else sc.ScriptError("Multiple state declarations not allowed");
bag.StateSet=true;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorRenderStyle (FScanner &sc, AActor *defaults, Baggage &bag)
{
static const char * renderstyles[]={
"NONE","NORMAL","FUZZY","SOULTRANS","OPTFUZZY","STENCIL","TRANSLUCENT", "ADD","SHADED", NULL};
static const int renderstyle_values[]={
STYLE_None, STYLE_Normal, STYLE_Fuzzy, STYLE_SoulTrans, STYLE_OptFuzzy,
STYLE_TranslucentStencil, STYLE_Translucent, STYLE_Add, STYLE_Shaded};
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->RenderStyle = LegacyRenderStyles[renderstyle_values[sc.MustMatchString(renderstyles)]];
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorAlpha (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString("DEFAULT"))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->alpha = gameinfo.gametype == GAME_Heretic ? HR_SHADOW : HX_SHADOW;
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->alpha=fixed_t(sc.Float*FRACUNIT);
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorObituary (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaString (AMETA_Obituary, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorHitObituary (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaString (AMETA_HitObituary, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDontHurtShooter (FScanner &sc, AActor *defaults, Baggage &bag)
{
bag.Info->Class->Meta.SetMetaInt (ACMETA_DontHurtShooter, true);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorExplosionRadius (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (ACMETA_ExplosionRadius, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorExplosionDamage (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (ACMETA_ExplosionDamage, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDeathHeight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
fixed_t h = fixed_t(sc.Float * FRACUNIT);
// AActor::Die() uses a height of 0 to mean "cut the height to 1/4",
// so if a height of 0 is desired, store it as -1.
bag.Info->Class->Meta.SetMetaFixed (AMETA_DeathHeight, h <= 0 ? -1 : h);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorBurnHeight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
fixed_t h = fixed_t(sc.Float * FRACUNIT);
// The note above for AMETA_DeathHeight also applies here.
bag.Info->Class->Meta.SetMetaFixed (AMETA_BurnHeight, h <= 0 ? -1 : h);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMaxTargetRange (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->maxtargetrange = fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMeleeThreshold (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->meleethreshold = fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMeleeDamage (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (ACMETA_MeleeDamage, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMeleeRange (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->meleerange = fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMeleeSound (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaInt (ACMETA_MeleeSound, S_FindSound(sc.String));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMissileType (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaInt (ACMETA_MissileName, FName(sc.String));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMissileHeight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
bag.Info->Class->Meta.SetMetaFixed (ACMETA_MissileHeight, fixed_t(sc.Float*FRACUNIT));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorTranslation (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckNumber())
{
int max = (gameinfo.gametype==GAME_Strife || (bag.Info->GameFilter&GAME_Strife)) ? 6:2;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.Number < 0 || sc.Number > max)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("Translation must be in the range [0,%d]", max);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->Translation = TRANSLATION(TRANSLATION_Standard, sc.Number);
}
else if (sc.CheckString("Ice"))
{
defaults->Translation = TRANSLATION(TRANSLATION_Standard, 7);
}
else
{
CurrentTranslation.MakeIdentity();
do
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.GetString();
AddToTranslation(sc.String);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
while (sc.CheckString(","));
defaults->Translation = StoreTranslation (sc);
}
}
//==========================================================================
//
//==========================================================================
static void ActorStencilColor (FScanner &sc, AActor *defaults, Baggage &bag)
{
int r,g,b;
if (sc.CheckNumber())
{
sc.MustGetNumber();
r=clamp<int>(sc.Number, 0, 255);
sc.CheckString(",");
sc.MustGetNumber();
g=clamp<int>(sc.Number, 0, 255);
sc.CheckString(",");
sc.MustGetNumber();
b=clamp<int>(sc.Number, 0, 255);
}
else
{
sc.MustGetString();
int c = V_GetColor(NULL, sc.String);
r=RPART(c);
g=GPART(c);
b=BPART(c);
}
defaults->fillcolor = MAKERGB(r,g,b) | (ColorMatcher.Pick (r, g, b) << 24);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorBloodColor (FScanner &sc, AActor *defaults, Baggage &bag)
{
int r,g,b;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckNumber())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
r = clamp<int>(sc.Number, 0, 255);
sc.CheckString(",");
sc.MustGetNumber();
g = clamp<int>(sc.Number, 0, 255);
sc.CheckString(",");
sc.MustGetNumber();
b = clamp<int>(sc.Number, 0, 255);
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
int c = V_GetColor(NULL, sc.String);
r = RPART(c);
g = GPART(c);
b = BPART(c);
}
PalEntry pe = MAKERGB(r,g,b);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
pe.a = CreateBloodTranslation(sc, pe);
if (DWORD(pe) == 0)
{
// If black is the first color being created it will create a value of 0
// which stands for 'no translation'
// Using (1,1,1) instead of (0,0,0) won't be noticable.
pe = MAKERGB(1,1,1);
}
bag.Info->Class->Meta.SetMetaInt (AMETA_BloodColor, pe);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorBloodType (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
FName blood = sc.String;
// normal blood
bag.Info->Class->Meta.SetMetaInt (AMETA_BloodType, blood);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString(","))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
blood = sc.String;
}
// blood splatter
bag.Info->Class->Meta.SetMetaInt (AMETA_BloodType2, blood);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString(","))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
blood = sc.String;
}
// axe blood
bag.Info->Class->Meta.SetMetaInt (AMETA_BloodType3, blood);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorBounceFactor (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
defaults->bouncefactor = clamp<fixed_t>(fixed_t(sc.Float * FRACUNIT), 0, FRACUNIT);
}
//==========================================================================
//
//==========================================================================
static void ActorWallBounceFactor (FScanner &sc, AActor *defaults, Baggage &bag)
{
sc.MustGetFloat ();
defaults->wallbouncefactor = clamp<fixed_t>(fixed_t(sc.Float * FRACUNIT), 0, FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorBounceCount (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->bouncecount = sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMinMissileChance (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->MinMissileChance=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDamageType (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
if (sc.Compare("Normal")) defaults->DamageType = NAME_None;
else defaults->DamageType=sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDamageFactor (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
FName dmgType;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.Compare("Normal")) dmgType = NAME_None;
else dmgType=sc.String;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(',');
sc.MustGetFloat();
bag.Info->SetDamageFactor(dmgType, FLOAT2FIXED(sc.Float));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorDecal (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->DecalGenerator = (FDecalBase *)intptr_t(int(FName(sc.String)));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMaxStepHeight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->MaxStepHeight=sc.Number * FRACUNIT;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMaxDropoffHeight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->MaxDropOffHeight=sc.Number * FRACUNIT;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorPoisonDamage (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (AMETA_PoisonDamage, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorFastSpeed (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
bag.Info->Class->Meta.SetMetaFixed (AMETA_FastSpeed, fixed_t(sc.Float*FRACUNIT));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorRadiusDamageFactor (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
bag.Info->Class->Meta.SetMetaFixed (AMETA_RDFactor, fixed_t(sc.Float*FRACUNIT));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorCameraheight (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
bag.Info->Class->Meta.SetMetaFixed (AMETA_CameraHeight, fixed_t(sc.Float*FRACUNIT));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorVSpeed (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->momz = fixed_t(sc.Float*FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorGravity (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.Float < 0.f)
sc.ScriptError ("Gravity must not be negative.");
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->gravity = FLOAT2FIXED (sc.Float);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.Float == 0.f)
defaults->flags |= MF_NOGRAVITY;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorClearFlags (FScanner &sc, AActor *defaults, Baggage &bag)
{
defaults->flags=defaults->flags3=defaults->flags4=defaults->flags5=0;
defaults->flags2&=MF2_ARGSDEFINED; // this flag must not be cleared
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorMonster (FScanner &sc, AActor *defaults, Baggage &bag)
{
// sets the standard flag for a monster
defaults->flags|=MF_SHOOTABLE|MF_COUNTKILL|MF_SOLID;
defaults->flags2|=MF2_PUSHWALL|MF2_MCROSS|MF2_PASSMOBJ;
defaults->flags3|=MF3_ISMONSTER;
defaults->flags4|=MF4_CANUSEWALLS;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ActorProjectile (FScanner &sc, AActor *defaults, Baggage &bag)
{
// sets the standard flags for a projectile
defaults->flags|=MF_NOBLOCKMAP|MF_NOGRAVITY|MF_DROPOFF|MF_MISSILE;
defaults->flags2|=MF2_IMPACT|MF2_PCROSS|MF2_NOTELEPORT;
if (gameinfo.gametype&GAME_Raven) defaults->flags5|=MF5_BLOODSPLATTER;
}
//==========================================================================
//
// Special inventory properties
//
//==========================================================================
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void AmmoBackpackAmount (FScanner &sc, AAmmo *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->BackpackAmount=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void AmmoBackpackMaxAmount (FScanner &sc, AAmmo *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->BackpackMaxAmount=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void AmmoDropAmount (FScanner &sc, AAmmo *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt (AIMETA_DropAmount, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ArmorMaxSaveAmount (FScanner &sc, ABasicArmorBonus *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->MaxSaveAmount = sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ArmorMaxBonus (FScanner &sc, ABasicArmorBonus *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->BonusCount = sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ArmorMaxBonusMax (FScanner &sc, ABasicArmorBonus *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->BonusMax = sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ArmorSaveAmount (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
// Special case here because this property has to work for 2 unrelated classes
if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(ABasicArmorPickup)))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
((ABasicArmorPickup*)defaults)->SaveAmount=sc.Number;
}
else if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(ABasicArmorBonus)))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
((ABasicArmorBonus*)defaults)->SaveAmount=sc.Number;
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("\"%s\" requires an actor of type \"Armor\"\n", sc.String);
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void ArmorSavePercent (FScanner &sc, AActor *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
if (sc.Float<0.0f) sc.Float=0.0f;
if (sc.Float>100.0f) sc.Float=100.0f;
// Special case here because this property has to work for 2 unrelated classes
if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(ABasicArmorPickup)))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
((ABasicArmorPickup*)defaults)->SavePercent=fixed_t(sc.Float*FRACUNIT/100.0f);
}
else if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(ABasicArmorBonus)))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
((ABasicArmorBonus*)defaults)->SavePercent=fixed_t(sc.Float*FRACUNIT/100.0f);
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("\"%s\" requires an actor of type \"Armor\"\n", sc.String);
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryAmount (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->Amount=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryIcon (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->Icon = TexMan.AddPatch (sc.String);
if (!defaults->Icon.isValid())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->Icon = TexMan.AddPatch (sc.String, ns_sprites);
if (!defaults->Icon.isValid())
{
// Don't print warnings if the item is for another game or if this is a shareware IWAD.
// Strife's teaser doesn't contain all the icon graphics of the full game.
if ((bag.Info->GameFilter == GAME_Any || bag.Info->GameFilter & gameinfo.gametype) &&
!(gameinfo.flags&GI_SHAREWARE) && Wads.GetLumpFile(sc.LumpNum) != 0)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
Printf("Icon '%s' for '%s' not found\n", sc.String, bag.Info->Class->TypeName.GetChars());
}
}
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryMaxAmount (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->MaxAmount=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryDefMaxAmount (FScanner &sc, AInventory *defaults, Baggage &bag)
{
defaults->MaxAmount = gameinfo.gametype == GAME_Heretic ? 16 : 25;
}
//==========================================================================
//
//==========================================================================
static void InventoryPickupflash (FScanner &sc, AInventory *defaults, Baggage &bag)
{
sc.MustGetString();
defaults->PickupFlash = fuglyname(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryPickupmsg (FScanner &sc, AInventory *defaults, Baggage &bag)
{
// allow game specific pickup messages
const char * games[] = {"Doom", "Heretic", "Hexen", "Raven", "Strife", "Chex", NULL};
int gamemode[]={GAME_Doom, GAME_Heretic, GAME_Hexen, GAME_Raven, GAME_Strife, GAME_Chex};
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
int game = sc.MatchString(games);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (game!=-1 && sc.CheckString(","))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
if (!(gameinfo.gametype&gamemode[game])) return;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
bag.Info->Class->Meta.SetMetaString(AIMETA_PickupMessage, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryPickupsound (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->PickupSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryRespawntics (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->RespawnTics=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryUsesound (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->UseSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void InventoryGiveQuest (FScanner &sc, AInventory *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt(AIMETA_GiveQuest, sc.Number);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void HealthLowMessage (FScanner &sc, AHealth *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
bag.Info->Class->Meta.SetMetaInt(AIMETA_LowHealth, sc.Number);
sc.MustGetStringName(",");
sc.MustGetString();
bag.Info->Class->Meta.SetMetaString(AIMETA_LowHealthMessage, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PuzzleitemNumber (FScanner &sc, APuzzleItem *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->PuzzleItemNumber=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PuzzleitemFailMsg (FScanner &sc, APuzzleItem *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
bag.Info->Class->Meta.SetMetaString(AIMETA_PuzzFailMessage, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponAmmoGive1 (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->AmmoGive1=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponAmmoGive2 (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->AmmoGive2=sc.Number;
}
//==========================================================================
//
// Passing these parameters is really tricky to allow proper inheritance
// and forward declarations. Here only a name is
// stored which must be resolved after everything has been declared
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponAmmoType1 (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->AmmoType1 = fuglyname(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponAmmoType2 (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->AmmoType2 = fuglyname(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponAmmoUse1 (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->AmmoUse1=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponAmmoUse2 (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->AmmoUse2=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponKickback (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->Kickback=sc.Number;
}
//==========================================================================
//
//==========================================================================
static void WeaponDefKickback (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
defaults->Kickback = gameinfo.defKickback;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponReadySound (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->ReadySound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponSelectionOrder (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->SelectionOrder=sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponSisterWeapon (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->SisterWeaponType=fuglyname(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponUpSound (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->UpSound = sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WeaponYAdjust (FScanner &sc, AWeapon *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat();
defaults->YAdjust=fixed_t(sc.Float * FRACUNIT);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WPieceValue (FScanner &sc, AWeaponPiece *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber();
defaults->PieceValue = 1 << (sc.Number-1);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void WPieceWeapon (FScanner &sc, AWeaponPiece *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->WeaponClass = fuglyname(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PowerupColor (FScanner &sc, APowerupGiver *defaults, Baggage &bag)
{
int r;
int g;
int b;
int alpha;
PalEntry * pBlendColor;
if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(APowerup)))
{
pBlendColor = &((APowerup*)defaults)->BlendColor;
}
else if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(APowerupGiver)))
{
pBlendColor = &((APowerupGiver*)defaults)->BlendColor;
}
else
{
sc.ScriptError("\"%s\" requires an actor of type \"Powerup\"\n", sc.String);
return;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckNumber())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
r=clamp<int>(sc.Number, 0, 255);
sc.CheckString(",");
sc.MustGetNumber();
g=clamp<int>(sc.Number, 0, 255);
sc.CheckString(",");
sc.MustGetNumber();
b=clamp<int>(sc.Number, 0, 255);
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.Compare("INVERSEMAP"))
{
defaults->BlendColor = INVERSECOLOR;
return;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare("GOLDMAP"))
{
defaults->BlendColor = GOLDCOLOR;
return;
}
// [BC] Yay, more hacks.
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if ( sc.Compare( "REDMAP" ))
{
defaults->BlendColor = REDCOLOR;
return;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if ( sc.Compare( "GREENMAP" ))
{
defaults->BlendColor = GREENCOLOR;
return;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int c = V_GetColor(NULL, sc.String);
r=RPART(c);
g=GPART(c);
b=BPART(c);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.CheckString(",");
sc.MustGetFloat();
alpha=int(sc.Float*255);
alpha=clamp<int>(alpha, 0, 255);
if (alpha!=0) *pBlendColor = MAKEARGB(alpha, r, g, b);
else *pBlendColor = 0;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PowerupDuration (FScanner &sc, APowerupGiver *defaults, Baggage &bag)
{
int *pEffectTics;
if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(APowerup)))
{
pEffectTics = &((APowerup*)defaults)->EffectTics;
}
else if (bag.Info->Class->IsDescendantOf(RUNTIME_CLASS(APowerupGiver)))
{
pEffectTics = &((APowerupGiver*)defaults)->EffectTics;
}
else
{
sc.ScriptError("\"%s\" requires an actor of type \"Powerup\"\n", sc.String);
return;
}
sc.MustGetNumber();
*pEffectTics = (sc.Number >= 0) ? sc.Number : -sc.Number * TICRATE;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PowerupMode (FScanner &sc, APowerupGiver *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->mode = (FName)sc.String;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PowerupType (FScanner &sc, APowerupGiver *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
defaults->PowerupType = fuglyname(sc.String);
}
//==========================================================================
//
// [GRB] Special player properties
//
//==========================================================================
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerDisplayName (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
bag.Info->Class->Meta.SetMetaString (APMETA_DisplayName, sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerSoundClass (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
FString tmp;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
tmp = sc.String;
tmp.ReplaceChars (' ', '_');
bag.Info->Class->Meta.SetMetaString (APMETA_SoundClass, tmp);
}
//==========================================================================
//
//==========================================================================
static void PlayerFace (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
FString tmp;
sc.MustGetString ();
tmp = sc.String;
if (tmp.Len() != 3)
{
Printf("Invalid face '%s' for '%s';\nSTF replacement codes must be 3 characters.\n",
sc.String, bag.Info->Class->TypeName.GetChars ());
}
tmp.ToUpper();
bool valid = (
(((tmp[0] >= 'A') && (tmp[0] <= 'Z')) || ((tmp[0] >= '0') && (tmp[0] <= '9'))) &&
(((tmp[1] >= 'A') && (tmp[1] <= 'Z')) || ((tmp[1] >= '0') && (tmp[1] <= '9'))) &&
(((tmp[2] >= 'A') && (tmp[2] <= 'Z')) || ((tmp[2] >= '0') && (tmp[2] <= '9')))
);
if (!valid)
{
Printf("Invalid face '%s' for '%s';\nSTF replacement codes must be alphanumeric.\n",
sc.String, bag.Info->Class->TypeName.GetChars ());
}
bag.Info->Class->Meta.SetMetaString (APMETA_Face, tmp);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerColorRange (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
int start, end;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
start = sc.Number;
sc.CheckString(",");
sc.MustGetNumber ();
end = sc.Number;
if (start > end)
swap (start, end);
bag.Info->Class->Meta.SetMetaInt (APMETA_ColorRange, (start & 255) | ((end & 255) << 8));
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerAttackZOffset (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
defaults->AttackZOffset = FLOAT2FIXED (sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerJumpZ (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
defaults->JumpZ = FLOAT2FIXED (sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerSpawnClass (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
if (sc.Compare ("Any"))
defaults->SpawnMask = 0;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Fighter"))
defaults->SpawnMask |= 1;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Cleric"))
defaults->SpawnMask |= 2;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.Compare ("Mage"))
defaults->SpawnMask |= 4;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (IsNum(sc.String))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int val = strtol(sc.String, NULL, 0);
if (val > 0) defaults->SpawnMask |= 1<<(val-1);
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerViewHeight (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
defaults->ViewHeight = FLOAT2FIXED (sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerForwardMove (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
defaults->ForwardMove1 = defaults->ForwardMove2 = FLOAT2FIXED (sc.Float);
if (CheckFloatParm (sc))
defaults->ForwardMove2 = FLOAT2FIXED (sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerSideMove (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
defaults->SideMove1 = defaults->SideMove2 = FLOAT2FIXED (sc.Float);
if (CheckFloatParm (sc))
defaults->SideMove2 = FLOAT2FIXED (sc.Float);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerMaxHealth (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->MaxHealth = sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerRunHealth (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetNumber ();
defaults->RunHealth = sc.Number;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerMorphWeapon (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
defaults->MorphWeapon = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerScoreIcon (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
defaults->ScoreIcon = TexMan.AddPatch (sc.String);
if (!defaults->ScoreIcon.isValid())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
defaults->ScoreIcon = TexMan.AddPatch (sc.String, ns_sprites);
if (!defaults->ScoreIcon.isValid())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
Printf("Icon '%s' for '%s' not found\n", sc.String, bag.Info->Class->TypeName.GetChars ());
}
}
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerCrouchSprite (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
for (int i = 0; i < sc.StringLen; i++)
{
sc.String[i] = toupper (sc.String[i]);
}
defaults->crouchsprite = GetSpriteIndex (sc.String);
}
//==========================================================================
//
//==========================================================================
static void PlayerDmgScreenColor (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
if (sc.CheckNumber ())
{
sc.MustGetNumber ();
defaults->RedDamageFade = clamp <float> (sc.Number, 0, 255);
sc.CheckString (",");
sc.MustGetNumber ();
defaults->GreenDamageFade = clamp <float> (sc.Number, 0, 255);
sc.CheckString (",");
sc.MustGetNumber ();
defaults->BlueDamageFade = clamp <float> (sc.Number, 0, 255);
}
else
{
sc.MustGetString ();
int c = V_GetColor (NULL, sc.String);
defaults->RedDamageFade = RPART (c);
defaults->GreenDamageFade = GPART (c);
defaults->BlueDamageFade = BPART (c);
}
}
//==========================================================================
//
// [GRB] Store start items in drop item list
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerStartItem (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
// create a linked list of dropitems
if (!bag.DropItemSet)
{
bag.DropItemSet = true;
bag.DropItemList = NULL;
}
FDropItem * di=new FDropItem;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString();
di->Name = sc.String;
di->probability = 255;
di->amount = 1;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (CheckNumParm(sc))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
di->amount = sc.Number;
}
di->Next = bag.DropItemList;
bag.DropItemList = di;
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerInvulMode (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
bag.Info->Class->Meta.SetMetaInt (APMETA_InvulMode, (FName)sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerHealRadius (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
bag.Info->Class->Meta.SetMetaInt (APMETA_HealingRadius, (FName)sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void PlayerHexenArmor (FScanner &sc, APlayerPawn *defaults, Baggage &bag)
{
for (int i=0;i<5;i++)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetFloat ();
bag.Info->Class->Meta.SetMetaFixed (APMETA_Hexenarmor0+i, FLOAT2FIXED (sc.Float));
if (i!=4) sc.MustGetStringName(",");
}
}
//==========================================================================
//
//==========================================================================
static void EggFXPlayerClass (FScanner &sc, AMorphProjectile *defaults, Baggage &bag)
{
sc.MustGetString ();
defaults->PlayerClass = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static void EggFXMonsterClass (FScanner &sc, AMorphProjectile *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
defaults->MonsterClass = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
static void EggFXDuration (FScanner &sc, AMorphProjectile *defaults, Baggage &bag)
{
sc.MustGetNumber ();
defaults->Duration = sc.Number;
}
//==========================================================================
//
//==========================================================================
static void EggFXMorphStyle (FScanner &sc, AMorphProjectile *defaults, Baggage &bag)
{
defaults->MorphStyle = ParseMorphStyle(sc);
}
//==========================================================================
//
//==========================================================================
static void EggFXMorphFlash (FScanner &sc, AMorphProjectile *defaults, Baggage &bag)
{
sc.MustGetString ();
defaults->MorphFlash = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
static void EggFXUnMorphFlash (FScanner &sc, AMorphProjectile *defaults, Baggage &bag)
{
sc.MustGetString ();
defaults->UnMorphFlash = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
static void PowerMorphPlayerClass (FScanner &sc, APowerMorph *defaults, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
defaults->PlayerClass = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
static void PowerMorphMorphStyle (FScanner &sc, APowerMorph *defaults, Baggage &bag)
{
defaults->MorphStyle = ParseMorphStyle(sc);
}
//==========================================================================
//
//==========================================================================
static void PowerMorphMorphFlash (FScanner &sc, APowerMorph *defaults, Baggage &bag)
{
sc.MustGetString ();
defaults->MorphFlash = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
static void PowerMorphUnMorphFlash (FScanner &sc, APowerMorph *defaults, Baggage &bag)
{
sc.MustGetString ();
defaults->UnMorphFlash = FName(sc.String);
}
//==========================================================================
//
//==========================================================================
static const ActorProps *APropSearch (const char *str, const ActorProps *props, int numprops)
{
int min = 0, max = numprops - 1;
while (min <= max)
{
int mid = (min + max) / 2;
int lexval = strcmp (str, props[mid].name);
if (lexval == 0)
{
return &props[mid];
}
else if (lexval > 0)
{
min = mid + 1;
}
else
{
max = mid - 1;
}
}
return NULL;
}
//==========================================================================
//
// all actor properties
//
//==========================================================================
#define apf ActorPropFunction
static const ActorProps props[] =
{
{ "activesound", ActorActiveSound, RUNTIME_CLASS(AActor) },
{ "alpha", ActorAlpha, RUNTIME_CLASS(AActor) },
{ "ammo.backpackamount", (apf)AmmoBackpackAmount, RUNTIME_CLASS(AAmmo) },
{ "ammo.backpackmaxamount", (apf)AmmoBackpackMaxAmount, RUNTIME_CLASS(AAmmo) },
{ "ammo.dropamount", (apf)AmmoDropAmount, RUNTIME_CLASS(AAmmo) },
{ "args", ActorArgs, RUNTIME_CLASS(AActor) },
{ "armor.maxbonus", (apf)ArmorMaxBonus, RUNTIME_CLASS(ABasicArmorBonus) },
{ "armor.maxbonusmax", (apf)ArmorMaxBonusMax, RUNTIME_CLASS(ABasicArmorBonus) },
{ "armor.maxsaveamount", (apf)ArmorMaxSaveAmount, RUNTIME_CLASS(ABasicArmorBonus) },
{ "armor.saveamount", (apf)ArmorSaveAmount, RUNTIME_CLASS(AActor) },
{ "armor.savepercent", (apf)ArmorSavePercent, RUNTIME_CLASS(AActor) },
{ "attacksound", ActorAttackSound, RUNTIME_CLASS(AActor) },
{ "bloodcolor", ActorBloodColor, RUNTIME_CLASS(AActor) },
{ "bloodtype", ActorBloodType, RUNTIME_CLASS(AActor) },
{ "bouncecount", ActorBounceCount, RUNTIME_CLASS(AActor) },
{ "bouncefactor", ActorBounceFactor, RUNTIME_CLASS(AActor) },
{ "burn", ActorBurnState, RUNTIME_CLASS(AActor) },
{ "burnheight", ActorBurnHeight, RUNTIME_CLASS(AActor) },
{ "cameraheight", ActorCameraheight, RUNTIME_CLASS(AActor) },
{ "clearflags", ActorClearFlags, RUNTIME_CLASS(AActor) },
{ "conversationid", ActorConversationID, RUNTIME_CLASS(AActor) },
{ "crash", ActorCrashState, RUNTIME_CLASS(AActor) },
{ "crush", ActorCrushState, RUNTIME_CLASS(AActor) },
{ "damage", ActorDamage, RUNTIME_CLASS(AActor) },
{ "damagefactor", ActorDamageFactor, RUNTIME_CLASS(AActor) },
{ "damagetype", ActorDamageType, RUNTIME_CLASS(AActor) },
{ "death", ActorDeathState, RUNTIME_CLASS(AActor) },
{ "deathheight", ActorDeathHeight, RUNTIME_CLASS(AActor) },
{ "deathsound", ActorDeathSound, RUNTIME_CLASS(AActor) },
{ "decal", ActorDecal, RUNTIME_CLASS(AActor) },
{ "disintegrate", ActorDisintegrateState, RUNTIME_CLASS(AActor) },
{ "donthurtshooter", ActorDontHurtShooter, RUNTIME_CLASS(AActor) },
{ "dropitem", ActorDropItem, RUNTIME_CLASS(AActor) },
{ "explosiondamage", ActorExplosionDamage, RUNTIME_CLASS(AActor) },
{ "explosionradius", ActorExplosionRadius, RUNTIME_CLASS(AActor) },
{ "fastspeed", ActorFastSpeed, RUNTIME_CLASS(AActor) },
{ "floatspeed", ActorFloatSpeed, RUNTIME_CLASS(AActor) },
{ "game", ActorGame, RUNTIME_CLASS(AActor) },
{ "gibhealth", ActorGibHealth, RUNTIME_CLASS(AActor) },
{ "gravity", ActorGravity, RUNTIME_CLASS(AActor) },
{ "heal", ActorHealState, RUNTIME_CLASS(AActor) },
{ "health", ActorHealth, RUNTIME_CLASS(AActor) },
{ "health.lowmessage", (apf)HealthLowMessage, RUNTIME_CLASS(AHealth) },
{ "height", ActorHeight, RUNTIME_CLASS(AActor) },
{ "hitobituary", ActorHitObituary, RUNTIME_CLASS(AActor) },
{ "howlsound", ActorHowlSound, RUNTIME_CLASS(AActor) },
{ "ice", ActorIceState, RUNTIME_CLASS(AActor) },
{ "inventory.amount", (apf)InventoryAmount, RUNTIME_CLASS(AInventory) },
{ "inventory.defmaxamount", (apf)InventoryDefMaxAmount, RUNTIME_CLASS(AInventory) },
{ "inventory.givequest", (apf)InventoryGiveQuest, RUNTIME_CLASS(AInventory) },
{ "inventory.icon", (apf)InventoryIcon, RUNTIME_CLASS(AInventory) },
{ "inventory.maxamount", (apf)InventoryMaxAmount, RUNTIME_CLASS(AInventory) },
{ "inventory.pickupflash", (apf)InventoryPickupflash, RUNTIME_CLASS(AInventory) },
{ "inventory.pickupmessage", (apf)InventoryPickupmsg, RUNTIME_CLASS(AInventory) },
{ "inventory.pickupsound", (apf)InventoryPickupsound, RUNTIME_CLASS(AInventory) },
{ "inventory.respawntics", (apf)InventoryRespawntics, RUNTIME_CLASS(AInventory) },
{ "inventory.usesound", (apf)InventoryUsesound, RUNTIME_CLASS(AInventory) },
{ "mass", ActorMass, RUNTIME_CLASS(AActor) },
{ "maxdropoffheight", ActorMaxDropoffHeight, RUNTIME_CLASS(AActor) },
{ "maxstepheight", ActorMaxStepHeight, RUNTIME_CLASS(AActor) },
{ "maxtargetrange", ActorMaxTargetRange, RUNTIME_CLASS(AActor) },
{ "melee", ActorMeleeState, RUNTIME_CLASS(AActor) },
{ "meleedamage", ActorMeleeDamage, RUNTIME_CLASS(AActor) },
{ "meleerange", ActorMeleeRange, RUNTIME_CLASS(AActor) },
{ "meleesound", ActorMeleeSound, RUNTIME_CLASS(AActor) },
{ "meleethreshold", ActorMeleeThreshold, RUNTIME_CLASS(AActor) },
{ "minmissilechance", ActorMinMissileChance, RUNTIME_CLASS(AActor) },
{ "missile", ActorMissileState, RUNTIME_CLASS(AActor) },
{ "missileheight", ActorMissileHeight, RUNTIME_CLASS(AActor) },
{ "missiletype", ActorMissileType, RUNTIME_CLASS(AActor) },
{ "monster", ActorMonster, RUNTIME_CLASS(AActor) },
{ "morphprojectile.duration", (apf)EggFXDuration, RUNTIME_CLASS(AMorphProjectile) },
{ "morphprojectile.monsterclass", (apf)EggFXMonsterClass, RUNTIME_CLASS(AMorphProjectile) },
{ "morphprojectile.morphflash", (apf)EggFXMorphFlash, RUNTIME_CLASS(AMorphProjectile) },
{ "morphprojectile.morphstyle", (apf)EggFXMorphStyle, RUNTIME_CLASS(AMorphProjectile) },
{ "morphprojectile.playerclass", (apf)EggFXPlayerClass, RUNTIME_CLASS(AMorphProjectile) },
{ "morphprojectile.unmorphflash", (apf)EggFXUnMorphFlash, RUNTIME_CLASS(AMorphProjectile) },
{ "obituary", ActorObituary, RUNTIME_CLASS(AActor) },
{ "pain", ActorPainState, RUNTIME_CLASS(AActor) },
{ "painchance", ActorPainChance, RUNTIME_CLASS(AActor) },
{ "painsound", ActorPainSound, RUNTIME_CLASS(AActor) },
{ "player.attackzoffset", (apf)PlayerAttackZOffset, RUNTIME_CLASS(APlayerPawn) },
{ "player.colorrange", (apf)PlayerColorRange, RUNTIME_CLASS(APlayerPawn) },
{ "player.crouchsprite", (apf)PlayerCrouchSprite, RUNTIME_CLASS(APlayerPawn) },
{ "player.damagescreencolor", (apf)PlayerDmgScreenColor, RUNTIME_CLASS(APlayerPawn) },
{ "player.displayname", (apf)PlayerDisplayName, RUNTIME_CLASS(APlayerPawn) },
{ "player.face", (apf)PlayerFace, RUNTIME_CLASS(APlayerPawn) },
{ "player.forwardmove", (apf)PlayerForwardMove, RUNTIME_CLASS(APlayerPawn) },
{ "player.healradiustype", (apf)PlayerHealRadius, RUNTIME_CLASS(APlayerPawn) },
{ "player.hexenarmor", (apf)PlayerHexenArmor, RUNTIME_CLASS(APlayerPawn) },
{ "player.invulnerabilitymode", (apf)PlayerInvulMode, RUNTIME_CLASS(APlayerPawn) },
{ "player.jumpz", (apf)PlayerJumpZ, RUNTIME_CLASS(APlayerPawn) },
{ "player.maxhealth", (apf)PlayerMaxHealth, RUNTIME_CLASS(APlayerPawn) },
{ "player.morphweapon", (apf)PlayerMorphWeapon, RUNTIME_CLASS(APlayerPawn) },
{ "player.runhealth", (apf)PlayerRunHealth, RUNTIME_CLASS(APlayerPawn) },
{ "player.scoreicon", (apf)PlayerScoreIcon, RUNTIME_CLASS(APlayerPawn) },
{ "player.sidemove", (apf)PlayerSideMove, RUNTIME_CLASS(APlayerPawn) },
{ "player.soundclass", (apf)PlayerSoundClass, RUNTIME_CLASS(APlayerPawn) },
{ "player.spawnclass", (apf)PlayerSpawnClass, RUNTIME_CLASS(APlayerPawn) },
{ "player.startitem", (apf)PlayerStartItem, RUNTIME_CLASS(APlayerPawn) },
{ "player.viewheight", (apf)PlayerViewHeight, RUNTIME_CLASS(APlayerPawn) },
{ "poisondamage", ActorPoisonDamage, RUNTIME_CLASS(AActor) },
{ "powermorph.morphflash", (apf)PowerMorphMorphFlash, RUNTIME_CLASS(APowerMorph) },
{ "powermorph.morphstyle", (apf)PowerMorphMorphStyle, RUNTIME_CLASS(APowerMorph) },
{ "powermorph.playerclass", (apf)PowerMorphPlayerClass, RUNTIME_CLASS(APowerMorph) },
{ "powermorph.unmorphflash", (apf)PowerMorphUnMorphFlash, RUNTIME_CLASS(APowerMorph) },
{ "powerup.color", (apf)PowerupColor, RUNTIME_CLASS(AInventory) },
{ "powerup.duration", (apf)PowerupDuration, RUNTIME_CLASS(AInventory) },
{ "powerup.mode", (apf)PowerupMode, RUNTIME_CLASS(APowerupGiver) },
{ "powerup.type", (apf)PowerupType, RUNTIME_CLASS(APowerupGiver) },
{ "projectile", ActorProjectile, RUNTIME_CLASS(AActor) },
{ "puzzleitem.failmessage", (apf)PuzzleitemFailMsg, RUNTIME_CLASS(APuzzleItem) },
{ "puzzleitem.number", (apf)PuzzleitemNumber, RUNTIME_CLASS(APuzzleItem) },
{ "radius", ActorRadius, RUNTIME_CLASS(AActor) },
{ "radiusdamagefactor", ActorRadiusDamageFactor, RUNTIME_CLASS(AActor) },
{ "raise", ActorRaiseState, RUNTIME_CLASS(AActor) },
{ "reactiontime", ActorReactionTime, RUNTIME_CLASS(AActor) },
{ "renderstyle", ActorRenderStyle, RUNTIME_CLASS(AActor) },
{ "scale", ActorScale, RUNTIME_CLASS(AActor) },
{ "see", ActorSeeState, RUNTIME_CLASS(AActor) },
{ "seesound", ActorSeeSound, RUNTIME_CLASS(AActor) },
{ "skip_super", ActorSkipSuper, RUNTIME_CLASS(AActor) },
{ "spawn", ActorSpawnState, RUNTIME_CLASS(AActor) },
{ "spawnid", ActorSpawnID, RUNTIME_CLASS(AActor) },
{ "speed", ActorSpeed, RUNTIME_CLASS(AActor) },
{ "states", ActorStates, RUNTIME_CLASS(AActor) },
{ "stencilcolor", ActorStencilColor, RUNTIME_CLASS(AActor) },
{ "tag", ActorTag, RUNTIME_CLASS(AActor) },
{ "translation", ActorTranslation, RUNTIME_CLASS(AActor) },
{ "vspeed", ActorVSpeed, RUNTIME_CLASS(AActor) },
{ "wallbouncefactor", ActorWallBounceFactor, RUNTIME_CLASS(AActor) },
{ "weapon.ammogive", (apf)WeaponAmmoGive1, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammogive1", (apf)WeaponAmmoGive1, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammogive2", (apf)WeaponAmmoGive2, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammotype", (apf)WeaponAmmoType1, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammotype1", (apf)WeaponAmmoType1, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammotype2", (apf)WeaponAmmoType2, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammouse", (apf)WeaponAmmoUse1, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammouse1", (apf)WeaponAmmoUse1, RUNTIME_CLASS(AWeapon) },
{ "weapon.ammouse2", (apf)WeaponAmmoUse2, RUNTIME_CLASS(AWeapon) },
{ "weapon.defaultkickback", (apf)WeaponDefKickback, RUNTIME_CLASS(AWeapon) },
{ "weapon.kickback", (apf)WeaponKickback, RUNTIME_CLASS(AWeapon) },
{ "weapon.readysound", (apf)WeaponReadySound, RUNTIME_CLASS(AWeapon) },
{ "weapon.selectionorder", (apf)WeaponSelectionOrder, RUNTIME_CLASS(AWeapon) },
{ "weapon.sisterweapon", (apf)WeaponSisterWeapon, RUNTIME_CLASS(AWeapon) },
{ "weapon.upsound", (apf)WeaponUpSound, RUNTIME_CLASS(AWeapon) },
{ "weapon.yadjust", (apf)WeaponYAdjust, RUNTIME_CLASS(AWeapon) },
{ "weaponpiece.number", (apf)WPieceValue, RUNTIME_CLASS(AWeaponPiece) },
{ "weaponpiece.weapon", (apf)WPieceWeapon, RUNTIME_CLASS(AWeaponPiece) },
{ "wound", ActorWoundState, RUNTIME_CLASS(AActor) },
{ "woundhealth", ActorWoundHealth, RUNTIME_CLASS(AActor) },
{ "xdeath", ActorXDeathState, RUNTIME_CLASS(AActor) },
{ "xscale", ActorXScale, RUNTIME_CLASS(AActor) },
{ "yscale", ActorYScale, RUNTIME_CLASS(AActor) },
// AWeapon:MinAmmo1 and 2 are never used so there is no point in adding them here!
};
static const ActorProps *is_actorprop (const char *str)
{
return APropSearch (str, props, sizeof(props)/sizeof(ActorProps));
}
//==========================================================================
//
// Parses an actor property
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
void ParseActorProperty(FScanner &sc, Baggage &bag)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
strlwr (sc.String);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
FString propname = sc.String;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
if (sc.CheckString ("."))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetString ();
propname += '.';
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
strlwr (sc.String);
propname += sc.String;
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.UnGet ();
}
const ActorProps *prop = is_actorprop (propname.GetChars());
if (prop != NULL)
{
if (!bag.Info->Class->IsDescendantOf(prop->type))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("\"%s\" requires an actor of type \"%s\"\n", propname.GetChars(), prop->type->TypeName.GetChars());
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
prop->Handler (sc, (AActor *)bag.Info->Class->Defaults, bag);
}
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError("\"%s\" is an unknown actor property\n", propname.GetChars());
}
}
//==========================================================================
//
// Finalizes an actor definition
//
//==========================================================================
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
void FinishActor(FScanner &sc, FActorInfo *info, Baggage &bag)
{
AActor *defaults = (AActor*)info->Class->Defaults;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
FinishStates (sc, info, defaults, bag);
InstallStates (info, defaults);
if (bag.DropItemSet)
{
if (bag.DropItemList == NULL)
{
if (info->Class->Meta.GetMetaInt (ACMETA_DropItems) != 0)
{
info->Class->Meta.SetMetaInt (ACMETA_DropItems, 0);
}
}
else
{
info->Class->Meta.SetMetaInt (ACMETA_DropItems,
DropItemList.Push (bag.DropItemList) + 1);
}
}
if (info->Class->IsDescendantOf (RUNTIME_CLASS(AInventory)))
{
defaults->flags |= MF_SPECIAL;
}
}