gzdoom-gles/src/textures/animations.cpp

989 lines
27 KiB
C++
Raw Normal View History

/*
** r_anim.cpp
** Routines for handling texture animation.
**
**---------------------------------------------------------------------------
** Copyright 1998-2006 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.
**
** 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 "doomtype.h"
#include "cmdlib.h"
#include "i_system.h"
#include "r_sky.h"
#include "m_random.h"
#include "d_player.h"
#include "p_spec.h"
#include "sc_man.h"
#include "templates.h"
#include "w_wad.h"
#include "g_level.h"
#include "farchive.h"
// MACROS ------------------------------------------------------------------
// TYPES -------------------------------------------------------------------
// PRIVATE FUNCTION PROTOTYPES ---------------------------------------------
// PUBLIC DATA DEFINITIONS -------------------------------------------------
// PRIVATE DATA DEFINITIONS ------------------------------------------------
static FRandom pr_animatepictures ("AnimatePics");
// CODE --------------------------------------------------------------------
//==========================================================================
//
// FTextureManager :: AddAnim
//
// Adds a new animation to the array. If one with the same basepic as the
// new one already exists, it is replaced.
//
//==========================================================================
FAnimDef *FTextureManager::AddAnim (FAnimDef *anim)
{
// Search for existing duplicate.
for (unsigned int i = 0; i < mAnimations.Size(); ++i)
{
if (mAnimations[i]->BasePic == anim->BasePic)
{
// Found one!
free (mAnimations[i]);
mAnimations[i] = anim;
return anim;
}
}
// Didn't find one, so add it at the end.
mAnimations.Push (anim);
return anim;
}
//==========================================================================
//
// FTextureManager :: AddSimpleAnim
//
// Creates an animation with simple characteristics. This is used for
// original Doom (non-ANIMDEFS-style) animations and Build animations.
//
//==========================================================================
FAnimDef *FTextureManager::AddSimpleAnim (FTextureID picnum, int animcount, DWORD speedmin, DWORD speedrange)
{
if (AreTexturesCompatible(picnum, picnum + (animcount - 1)))
{
FAnimDef *anim = (FAnimDef *)M_Malloc (sizeof(FAnimDef));
anim->CurFrame = 0;
anim->BasePic = picnum;
anim->NumFrames = animcount;
anim->AnimType = FAnimDef::ANIM_Forward;
anim->bDiscrete = false;
anim->SwitchTime = 0;
anim->Frames[0].SpeedMin = speedmin;
anim->Frames[0].SpeedRange = speedrange;
anim->Frames[0].FramePic = anim->BasePic;
return AddAnim (anim);
}
return NULL;
}
//==========================================================================
//
// FTextureManager :: AddComplexAnim
//
// Creates an animation with individually defined frames.
//
//==========================================================================
FAnimDef *FTextureManager::AddComplexAnim (FTextureID picnum, const TArray<FAnimDef::FAnimFrame> &frames)
{
FAnimDef *anim = (FAnimDef *)M_Malloc (sizeof(FAnimDef) + (frames.Size()-1) * sizeof(frames[0]));
anim->BasePic = picnum;
anim->NumFrames = frames.Size();
anim->CurFrame = 0;
anim->AnimType = FAnimDef::ANIM_Forward;
anim->bDiscrete = true;
anim->SwitchTime = 0;
memcpy (&anim->Frames[0], &frames[0], frames.Size() * sizeof(frames[0]));
return AddAnim (anim);
}
//==========================================================================
//
// FTextureManager :: Initanimated
//
// [description copied from BOOM]
// Load the table of animation definitions, checking for existence of
// the start and end of each frame. If the start doesn't exist the sequence
// is skipped, if the last doesn't exist, BOOM exits.
//
// Wall/Flat animation sequences, defined by name of first and last frame,
// The full animation sequence is given using all lumps between the start
// and end entry, in the order found in the WAD file.
//
// This routine modified to read its data from a predefined lump or
// PWAD lump called ANIMATED rather than a static table in this module to
// allow wad designers to insert or modify animation sequences.
//
// Lump format is an array of byte packed animdef_t structures, terminated
// by a structure with istexture == -1. The lump can be generated from a
// text source file using SWANTBLS.EXE, distributed with the BOOM utils.
// The standard list of switches and animations is contained in the example
// source text file DEFSWANI.DAT also in the BOOM util distribution.
//
// [RH] Rewritten to support BOOM ANIMATED lump but also make absolutely
// no assumptions about how the compiler packs the animdefs array.
//
//==========================================================================
CVAR(Bool, debuganimated, false, 0)
void FTextureManager::InitAnimated (void)
{
const BITFIELD texflags = TEXMAN_Overridable;
// I think better not! This is only for old ANIMATED definitions that
// don't know about ZDoom's more flexible texture system.
// | FTextureManager::TEXMAN_TryAny;
int lumpnum = Wads.CheckNumForName ("ANIMATED");
if (lumpnum != -1)
{
FMemLump animatedlump = Wads.ReadLump (lumpnum);
int animatedlen = Wads.LumpLength(lumpnum);
const char *animdefs = (const char *)animatedlump.GetMem();
const char *anim_p;
FTextureID pic1, pic2;
int animtype;
DWORD animspeed;
// Init animation
animtype = FAnimDef::ANIM_Forward;
for (anim_p = animdefs; *anim_p != -1; anim_p += 23)
{
// make sure the current chunk of data is inside the lump boundaries.
if (anim_p + 22 >= animdefs + animatedlen)
{
I_Error("Tried to read past end of ANIMATED lump.");
}
if (*anim_p /* .istexture */ & 1)
{
// different episode ?
if (!(pic1 = CheckForTexture (anim_p + 10 /* .startname */, FTexture::TEX_Wall, texflags)).Exists() ||
!(pic2 = CheckForTexture (anim_p + 1 /* .endname */, FTexture::TEX_Wall, texflags)).Exists())
continue;
// [RH] Bit 1 set means allow decals on walls with this texture
Texture(pic2)->bNoDecals = Texture(pic1)->bNoDecals = !(*anim_p & 2);
}
else
{
if (!(pic1 = CheckForTexture (anim_p + 10 /* .startname */, FTexture::TEX_Flat, texflags)).Exists() ||
!(pic2 = CheckForTexture (anim_p + 1 /* .startname */, FTexture::TEX_Flat, texflags)).Exists())
continue;
}
FTexture *tex1 = Texture(pic1);
FTexture *tex2 = Texture(pic2);
animspeed = (BYTE(anim_p[19]) << 0) | (BYTE(anim_p[20]) << 8) |
(BYTE(anim_p[21]) << 16) | (BYTE(anim_p[22]) << 24);
// SMMU-style swirly hack? Don't apply on already-warping texture
if (animspeed > 65535 && tex1 != NULL && !tex1->bWarped)
{
FTexture *warper = new FWarp2Texture (tex1);
ReplaceTexture (pic1, warper, false);
}
// These tests were not really relevant for swirling textures, or even potentially
// harmful, so they have been moved to the else block.
else
{
if (tex1->UseType != tex2->UseType)
{
// not the same type -
continue;
}
if (debuganimated)
{
Printf("Defining animation '%s' (texture %d, lump %d, file %d) to '%s' (texture %d, lump %d, file %d)\n",
tex1->Name.GetChars(), pic1.GetIndex(), tex1->GetSourceLump(), Wads.GetLumpFile(tex1->GetSourceLump()),
tex2->Name.GetChars(), pic2.GetIndex(), tex2->GetSourceLump(), Wads.GetLumpFile(tex2->GetSourceLump()));
}
if (pic1 == pic2)
{
// This animation only has one frame. Skip it. (Doom aborted instead.)
Printf ("Animation %s in ANIMATED has only one frame\n", anim_p + 10);
continue;
}
// [RH] Allow for backward animations as well as forward.
else if (pic1 > pic2)
{
swapvalues (pic1, pic2);
animtype = FAnimDef::ANIM_Backward;
}
// Speed is stored as tics, but we want ms so scale accordingly.
FAnimDef *adef = AddSimpleAnim (pic1, pic2 - pic1 + 1, Scale (animspeed, 1000, 35));
if (adef != NULL) adef->AnimType = animtype;
}
}
}
}
//==========================================================================
//
// FTextureManager :: InitAnimDefs
//
// This uses a Hexen ANIMDEFS lump to define the animation sequences
//
//==========================================================================
void FTextureManager::InitAnimDefs ()
{
const BITFIELD texflags = TEXMAN_Overridable | TEXMAN_TryAny;
int lump, lastlump = 0;
while ((lump = Wads.FindLump ("ANIMDEFS", &lastlump)) != -1)
{
FScanner sc(lump);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.GetString ())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("flat"))
{
ParseAnim (sc, FTexture::TEX_Flat);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("texture"))
{
ParseAnim (sc, FTexture::TEX_Wall);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("switch"))
{
ProcessSwitchDef (sc);
}
// [GRB] Added warping type 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 ("warp") || sc.Compare ("warp2"))
{
ParseWarp(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
else if (sc.Compare ("cameratexture"))
{
ParseCameraTexture(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
else if (sc.Compare ("animatedDoor"))
{
ParseAnimatedDoor (sc);
}
else if (sc.Compare("skyoffset"))
{
sc.MustGetString ();
FTextureID picnum = CheckForTexture (sc.String, FTexture::TEX_Wall, texflags);
sc.MustGetNumber();
if (picnum.Exists())
{
Texture(picnum)->SkyOffset = 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 (NULL);
}
}
}
}
//==========================================================================
//
// FTextureManager :: ParseAnim
//
// Parse a single animation definition out of an ANIMDEFS lump and
// create the corresponding animation structure.
//
//==========================================================================
void FTextureManager::ParseAnim (FScanner &sc, int usetype)
{
const BITFIELD texflags = TEXMAN_Overridable | TEXMAN_TryAny;
TArray<FAnimDef::FAnimFrame> frames (32);
FTextureID picnum;
int defined = 0;
bool optional = false, missing = false;
FAnimDef *ani = NULL;
BYTE type = FAnimDef::ANIM_Forward;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("optional"))
{
optional = 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
sc.MustGetString ();
}
picnum = CheckForTexture (sc.String, usetype, texflags);
if (!picnum.Exists())
{
if (optional)
{
missing = 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
Printf (PRINT_BOLD, "ANIMDEFS: Can't find %s\n", sc.String);
}
}
// no decals on animating textures, by default
if (picnum.isValid())
{
Texture(picnum)->bNoDecals = 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
while (sc.GetString ())
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("allowdecals"))
{
if (picnum.isValid())
{
Texture(picnum)->bNoDecals = false;
}
continue;
}
else if (sc.Compare ("Oscillate"))
{
if (type == FAnimDef::ANIM_Random)
{
sc.ScriptError ("You cannot use \"random\" and \"oscillate\" together in a single animation.");
}
type = FAnimDef::ANIM_OscillateUp;
}
else if (sc.Compare("Random"))
{
if (type == FAnimDef::ANIM_OscillateUp)
{
sc.ScriptError ("You cannot use \"random\" and \"oscillate\" together in a single animation.");
}
type = FAnimDef::ANIM_Random;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("range"))
{
if (defined == 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
sc.ScriptError ("You cannot use \"pic\" and \"range\" together in a single animation.");
}
if (defined == 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
sc.ScriptError ("You can only use one \"range\" per animation.");
}
defined = 1;
ani = ParseRangeAnim (sc, picnum, usetype, missing);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("pic"))
{
if (defined == 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
sc.ScriptError ("You cannot use \"pic\" and \"range\" together in a single animation.");
}
defined = 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
ParsePicAnim (sc, picnum, usetype, missing, frames);
}
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 ();
break;
}
}
// If base pic is not present, don't add this anim
// ParseRangeAnim adds the anim itself, but ParsePicAnim does not.
if (picnum.isValid() && defined == 2)
{
if (frames.Size() < 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
sc.ScriptError ("Animation needs at least 2 frames");
}
ani = AddComplexAnim (picnum, frames);
}
if (ani != NULL && type != FAnimDef::ANIM_Forward)
{
if (ani->AnimType == FAnimDef::ANIM_Backward && type == FAnimDef::ANIM_OscillateUp) ani->AnimType = FAnimDef::ANIM_OscillateDown;
else ani->AnimType = type;
}
}
//==========================================================================
//
// FTextureManager :: ParseRangeAnim
//
// Parse an animation defined using "range". Not that one range entry is
// enough to define a complete animation, unlike "pic".
//
//==========================================================================
FAnimDef *FTextureManager::ParseRangeAnim (FScanner &sc, FTextureID picnum, int usetype, bool missing)
{
int type;
FTextureID framenum;
DWORD min, max;
type = FAnimDef::ANIM_Forward;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
framenum = ParseFramenum (sc, picnum, usetype, missing);
ParseTime (sc, min, max);
if (framenum == picnum || !picnum.Exists())
{
return NULL; // Animation is only one frame or does not exist
}
if (framenum < picnum)
{
type = FAnimDef::ANIM_Backward;
Texture(framenum)->bNoDecals = Texture(picnum)->bNoDecals;
swapvalues (framenum, picnum);
}
FAnimDef *ani = AddSimpleAnim (picnum, framenum - picnum + 1, min, max - min);
2015-03-21 12:08:08 +00:00
if (ani != NULL) ani->AnimType = type;
return ani;
}
//==========================================================================
//
// FTextureManager :: ParsePicAnim
//
// Parse a single frame from ANIMDEFS defined using "pic".
//
//==========================================================================
void FTextureManager::ParsePicAnim (FScanner &sc, FTextureID picnum, int usetype, bool missing, TArray<FAnimDef::FAnimFrame> &frames)
{
FTextureID framenum;
DWORD min = 1, max = 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
framenum = ParseFramenum (sc, picnum, usetype, missing);
ParseTime (sc, min, max);
if (picnum.isValid())
{
FAnimDef::FAnimFrame frame;
frame.SpeedMin = min;
frame.SpeedRange = max - min;
frame.FramePic = framenum;
frames.Push (frame);
}
}
//==========================================================================
//
// FTextureManager :: ParseFramenum
//
// Reads a frame's texture from ANIMDEFS. It can either be an integral
// offset from basepicnum or a specific texture name.
//
//==========================================================================
FTextureID FTextureManager::ParseFramenum (FScanner &sc, FTextureID basepicnum, int usetype, bool allowMissing)
{
const BITFIELD texflags = TEXMAN_Overridable | TEXMAN_TryAny;
FTextureID framenum;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 (IsNum (sc.String))
{
framenum = basepicnum + (atoi(sc.String) - 1);
}
else
{
framenum = CheckForTexture (sc.String, usetype, texflags);
if (!framenum.Exists() && !allowMissing)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 texture %s", sc.String);
}
}
return framenum;
}
//==========================================================================
//
// FTextureManager :: ParseTime
//
// Reads a tics or rand time definition from ANIMDEFS.
//
//==========================================================================
void FTextureManager::ParseTime (FScanner &sc, DWORD &min, DWORD &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.MustGetString ();
if (sc.Compare ("tics"))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ();
min = max = DWORD(sc.Float * 1000 / 35);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("rand"))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ();
min = DWORD(sc.Float * 1000 / 35);
sc.MustGetFloat ();
max = DWORD(sc.Float * 1000 / 35);
}
else
{
min = max = 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
sc.ScriptError ("Must specify a duration for animation frame");
}
}
//==========================================================================
//
// FTextureManager :: ParseWarp
//
// Parses a warping texture definition
//
//==========================================================================
void FTextureManager::ParseWarp(FScanner &sc)
{
const BITFIELD texflags = TEXMAN_Overridable | TEXMAN_TryAny | TEXMAN_ShortNameOnly;
bool isflat = false;
bool type2 = sc.Compare ("warp2"); // [GRB]
sc.MustGetString ();
if (sc.Compare ("flat"))
{
isflat = true;
sc.MustGetString ();
}
else if (sc.Compare ("texture"))
{
isflat = false;
sc.MustGetString ();
}
else
{
sc.ScriptError (NULL);
}
FTextureID picnum = CheckForTexture (sc.String, isflat ? FTexture::TEX_Flat : FTexture::TEX_Wall, texflags);
if (picnum.isValid())
{
FTexture *warper = Texture(picnum);
// don't warp a texture more than once
if (!warper->bWarped)
{
if (type2) warper = new FWarp2Texture (warper);
else warper = new FWarpTexture (warper);
ReplaceTexture (picnum, warper, false);
}
if (sc.CheckFloat())
{
static_cast<FWarpTexture*>(warper)->SetSpeed(float(sc.Float));
}
// No decals on warping textures, by default.
// Warping information is taken from the last warp
// definition for this texture.
warper->bNoDecals = true;
if (sc.GetString ())
{
if (sc.Compare ("allowdecals"))
{
warper->bNoDecals = false;
}
else
{
sc.UnGet ();
}
}
}
}
//==========================================================================
//
// ParseCameraTexture
//
// Parses a camera texture definition
//
//==========================================================================
void FTextureManager::ParseCameraTexture(FScanner &sc)
{
const BITFIELD texflags = TEXMAN_Overridable | TEXMAN_TryAny | TEXMAN_ShortNameOnly;
int width, height;
int fitwidth, fitheight;
FString picname;
sc.MustGetString ();
picname = sc.String;
sc.MustGetNumber ();
width = sc.Number;
sc.MustGetNumber ();
height = sc.Number;
FTextureID picnum = CheckForTexture (picname, FTexture::TEX_Flat, texflags);
FTexture *viewer = new FCanvasTexture (picname, width, height);
if (picnum.Exists())
{
FTexture *oldtex = Texture(picnum);
fitwidth = oldtex->GetScaledWidth ();
fitheight = oldtex->GetScaledHeight ();
viewer->UseType = oldtex->UseType;
ReplaceTexture (picnum, viewer, true);
}
else
{
fitwidth = width;
fitheight = height;
// [GRB] No need for oldtex
viewer->UseType = FTexture::TEX_Wall;
AddTexture (viewer);
}
if (sc.GetString())
{
if (sc.Compare ("fit"))
{
sc.MustGetNumber ();
fitwidth = sc.Number;
sc.MustGetNumber ();
fitheight = sc.Number;
}
else
{
sc.UnGet ();
}
}
if (sc.GetString())
{
if (sc.Compare("WorldPanning"))
{
viewer->bWorldPanning = true;
}
else
{
sc.UnGet();
}
}
viewer->SetScaledSize(fitwidth, fitheight);
}
//==========================================================================
//
// FTextureManager :: FixAnimations
//
// Copy the "front sky" flag from an animated texture to the rest
// of the textures in the animation, and make every texture in an
// animation range use the same setting for bNoDecals.
//
//==========================================================================
void FTextureManager::FixAnimations ()
{
unsigned int i;
int j;
for (i = 0; i < mAnimations.Size(); ++i)
{
FAnimDef *anim = mAnimations[i];
if (anim->bDiscrete)
{
if (Texture(anim->BasePic)->bNoRemap0)
{
for (j = 0; j < anim->NumFrames; ++j)
{
Texture(anim->Frames[j].FramePic)->SetFrontSkyLayer ();
}
}
}
else
{
bool nodecals;
bool noremap = false;
const char *name;
name = Texture(anim->BasePic)->Name;
nodecals = Texture(anim->BasePic)->bNoDecals;
for (j = 0; j < anim->NumFrames; ++j)
{
FTexture *tex = Texture(anim->BasePic + j);
noremap |= tex->bNoRemap0;
tex->bNoDecals = nodecals;
}
if (noremap)
{
for (j = 0; j < anim->NumFrames; ++j)
{
Texture(anim->BasePic + j)->SetFrontSkyLayer ();
}
}
}
}
}
//==========================================================================
//
// ParseAnimatedDoor
//
// Parses an animated door definition
//
//==========================================================================
void FTextureManager::ParseAnimatedDoor(FScanner &sc)
{
const BITFIELD texflags = TEXMAN_Overridable | TEXMAN_TryAny;
FDoorAnimation anim;
TArray<FTextureID> frames;
bool error = false;
FTextureID v;
sc.MustGetString();
anim.BaseTexture = CheckForTexture (sc.String, FTexture::TEX_Wall, texflags);
if (!anim.BaseTexture.Exists())
{
error = true;
}
2013-12-01 20:11:33 +00:00
else
{
Texture(anim.BaseTexture)->bNoDecals = true;
}
while (sc.GetString())
{
if (sc.Compare ("opensound"))
{
sc.MustGetString ();
anim.OpenSound = sc.String;
}
else if (sc.Compare ("closesound"))
{
sc.MustGetString ();
anim.CloseSound = sc.String;
}
else if (sc.Compare ("pic"))
{
sc.MustGetString ();
if (IsNum (sc.String))
{
v = anim.BaseTexture + (atoi(sc.String) - 1);
}
else
{
v = CheckForTexture (sc.String, FTexture::TEX_Wall, texflags);
if (!v.Exists() && anim.BaseTexture.Exists() && !error)
{
sc.ScriptError ("Unknown texture %s", sc.String);
}
}
frames.Push(v);
}
else if (sc.Compare("allowdecals"))
{
2013-12-01 20:11:33 +00:00
if (anim.BaseTexture.Exists()) Texture(anim.BaseTexture)->bNoDecals = false;
}
else
{
sc.UnGet ();
break;
}
}
if (!error)
{
anim.TextureFrames = new FTextureID[frames.Size()];
memcpy (anim.TextureFrames, &frames[0], sizeof(FTextureID) * frames.Size());
anim.NumTextureFrames = frames.Size();
mAnimatedDoors.Push (anim);
}
}
//==========================================================================
//
// Return index into "DoorAnimations" array for which door type to use
//
//==========================================================================
FDoorAnimation *FTextureManager::FindAnimatedDoor (FTextureID picnum)
{
unsigned int i;
for (i = 0; i < mAnimatedDoors.Size(); ++i)
{
if (picnum == mAnimatedDoors[i].BaseTexture)
return &mAnimatedDoors[i];
}
return NULL;
}
//==========================================================================
//
// FAnimDef :: SetSwitchTime
//
// Determines when to switch to the next frame.
//
//==========================================================================
void FAnimDef::SetSwitchTime (DWORD mstime)
{
int speedframe = bDiscrete ? CurFrame : 0;
SwitchTime = mstime + Frames[speedframe].SpeedMin;
if (Frames[speedframe].SpeedRange != 0)
{
SwitchTime += pr_animatepictures(Frames[speedframe].SpeedRange);
}
}
//==========================================================================
//
// FTextureManager :: SetTranslation
//
// Sets animation translation for a texture
//
//==========================================================================
void FTextureManager::SetTranslation (FTextureID fromtexnum, FTextureID totexnum)
{
if ((size_t)fromtexnum.texnum < Translation.Size())
{
if ((size_t)totexnum.texnum >= Textures.Size())
{
totexnum.texnum = fromtexnum.texnum;
}
Translation[fromtexnum.texnum] = totexnum.texnum;
}
}
//==========================================================================
//
// FTextureManager :: UpdateAnimations
//
// Updates texture translations for each animation and scrolls the skies.
//
//==========================================================================
void FTextureManager::UpdateAnimations (DWORD mstime)
{
for (unsigned int j = 0; j < mAnimations.Size(); ++j)
{
FAnimDef *anim = mAnimations[j];
// If this is the first time through R_UpdateAnimations, just
// initialize the anim's switch time without actually animating.
if (anim->SwitchTime == 0)
{
anim->SetSwitchTime (mstime);
}
else while (anim->SwitchTime <= mstime)
{ // Multiple frames may have passed since the last time calling
// R_UpdateAnimations, so be sure to loop through them all.
switch (anim->AnimType)
{
default:
case FAnimDef::ANIM_Forward:
anim->CurFrame = (anim->CurFrame + 1) % anim->NumFrames;
break;
case FAnimDef::ANIM_Backward:
if (anim->CurFrame == 0)
{
anim->CurFrame = anim->NumFrames - 1;
}
else
{
anim->CurFrame -= 1;
}
break;
case FAnimDef::ANIM_Random:
// select a random frame other than the current one
if (anim->NumFrames > 1)
{
WORD rndFrame = (WORD)pr_animatepictures(anim->NumFrames - 1);
if (rndFrame >= anim->CurFrame) rndFrame++;
anim->CurFrame = rndFrame;
}
break;
case FAnimDef::ANIM_OscillateUp:
anim->CurFrame = anim->CurFrame + 1;
if (anim->CurFrame >= anim->NumFrames - 1)
{
anim->AnimType = FAnimDef::ANIM_OscillateDown;
}
break;
case FAnimDef::ANIM_OscillateDown:
anim->CurFrame = anim->CurFrame - 1;
if (anim->CurFrame == 0)
{
anim->AnimType = FAnimDef::ANIM_OscillateUp;
}
break;
}
anim->SetSwitchTime (mstime);
}
if (anim->bDiscrete)
{
SetTranslation (anim->BasePic, anim->Frames[anim->CurFrame].FramePic);
}
else
{
for (unsigned int i = 0; i < anim->NumFrames; i++)
{
SetTranslation (anim->BasePic + i, anim->BasePic + (i + anim->CurFrame) % anim->NumFrames);
}
}
}
}
//==========================================================================
//
// operator<<
//
//==========================================================================
template<> FArchive &operator<< (FArchive &arc, FDoorAnimation* &Doorani)
{
if (arc.IsStoring())
{
arc << Doorani->BaseTexture;
}
else
{
FTextureID tex;
arc << tex;
Doorani = TexMan.FindAnimatedDoor(tex);
}
return arc;
}