gzdoom-gles/src/thingdef/thingdef_exp.cpp

1616 lines
34 KiB
C++
Raw Normal View History

/*
** thingdef_exp.cpp
**
** Expression parsing / runtime evaluating support
**
**---------------------------------------------------------------------------
** Copyright 2005 Jan Cholasta
** All rights reserved.
**
** Redistribution and use in source and binary forms, with or without
** modification, are permitted provided that the following conditions
** are met:
**
** 1. Redistributions of source code must retain the above copyright
** notice, this list of conditions and the following disclaimer.
** 2. Redistributions in binary form must reproduce the above copyright
** notice, this list of conditions and the following disclaimer in the
** documentation and/or other materials provided with the distribution.
** 3. The name of the author may not be used to endorse or promote products
** derived from this software without specific prior written permission.
** 4. When not used as part of ZDoom or a ZDoom derivative, this code will be
** covered by the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or (at
** your option) any later version.
**
** THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
** IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
** OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
** IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
** INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
** NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
** DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
** (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
** THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
**---------------------------------------------------------------------------
**
*/
2006-04-16 03:31:40 +00:00
#include "actor.h"
2006-04-15 15:00:29 +00:00
#include "sc_man.h"
#include "tarray.h"
#include "templates.h"
#include "cmdlib.h"
#include "i_system.h"
#include "m_random.h"
#include "a_pickups.h"
#include "thingdef.h"
#include "p_lnspec.h"
void InitExpressions ();
void ClearExpressions ();
FRandom pr_exrandom ("EX_Random");
enum ExpOp
{
EX_NOP,
EX_Const,
EX_Var,
EX_Compl, // ~exp
EX_Not, // !exp
// EX_Plus, // +exp
EX_Minus, // -exp
EX_Mul, // exp * exp
EX_Div, // exp / exp
EX_Mod, // exp % exp
EX_Add, // exp + exp
EX_Sub, // exp - exp
EX_LShift, // exp << exp
EX_RShift, // exp >> exp
EX_LT, // exp < exp
EX_GT, // exp > exp
EX_LE, // exp <= exp
EX_GE, // exp >= exp
EX_Eq, // exp == exp
EX_NE, // exp != exp
EX_And, // exp & exp
EX_Xor, // exp ^ exp
EX_Or, // exp | exp
EX_LogAnd, // exp && exp
EX_LogOr, // exp || exp
EX_Cond, // exp ? exp : exp
EX_Random, // random (min, max)
EX_Random2, // random2 ([mask])
EX_Sin, // sin (angle)
EX_Cos, // cos (angle)
EX_ActionSpecial,
EX_Right,
};
enum ExpValType
{
VAL_Int,
VAL_Float,
};
struct ExpVal
{
ExpValType Type;
union
{
int Int;
float Float;
};
};
typedef ExpVal (*ExpVarGet) (AActor *, int);
ExpVal GetAlpha (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->alpha);
return val;
}
ExpVal GetAngle (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = (float)actor->angle / ANGLE_1;
return val;
}
ExpVal GetArgs (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Int;
val.Int = actor->args[id];
return val;
}
ExpVal GetCeilingZ (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->ceilingz);
return val;
}
ExpVal GetFloorZ (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->floorz);
return val;
}
ExpVal GetHealth (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Int;
val.Int = actor->health;
return val;
}
ExpVal GetPitch (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = (float)actor->pitch / ANGLE_1;
return val;
}
ExpVal GetSpecial (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Int;
val.Int = actor->special;
return val;
}
ExpVal GetTID (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Int;
val.Int = actor->tid;
return val;
}
ExpVal GetTIDToHate (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Int;
val.Int = actor->TIDtoHate;
return val;
}
ExpVal GetWaterLevel (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Int;
val.Int = actor->waterlevel;
return val;
}
ExpVal GetX (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->x);
return val;
}
ExpVal GetY (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->y);
return val;
}
ExpVal GetZ (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->z);
return val;
}
ExpVal GetMomX (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->momx);
return val;
}
ExpVal GetMomY (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->momy);
return val;
}
ExpVal GetMomZ (AActor *actor, int id)
{
ExpVal val;
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (actor->momz);
return val;
}
static struct FExpVar
{
ENamedName name; // identifier
int array; // array size (0 if not an array)
ExpVarGet get;
} ExpVars[] = {
{ NAME_Alpha, 0, GetAlpha },
{ NAME_Angle, 0, GetAngle },
{ NAME_Args, 5, GetArgs },
{ NAME_CeilingZ, 0, GetCeilingZ },
{ NAME_FloorZ, 0, GetFloorZ },
{ NAME_Health, 0, GetHealth },
{ NAME_Pitch, 0, GetPitch },
{ NAME_Special, 0, GetSpecial },
{ NAME_TID, 0, GetTID },
{ NAME_TIDtoHate, 0, GetTIDToHate },
{ NAME_WaterLevel, 0, GetWaterLevel },
{ NAME_X, 0, GetX },
{ NAME_Y, 0, GetY },
{ NAME_Z, 0, GetZ },
{ NAME_MomX, 0, GetMomX },
{ NAME_MomY, 0, GetMomY },
{ NAME_MomZ, 0, GetMomZ },
};
struct ExpData;
static ExpVal EvalExpression (ExpData *data, AActor *self, const PClass *cls);
2006-04-15 15:00:29 +00:00
struct ExpData
{
ExpData ()
{
Type = EX_NOP;
Value.Type = VAL_Int;
Value.Int = 0;
RNG = NULL;
for (int i = 0; i < 2; i++)
Children[i] = NULL;
}
~ExpData ()
{
for (int i = 0; i < 2; i++)
{
if (Children[i])
{
delete Children[i];
}
}
}
// Try to evaluate constant expression
void EvalConst (const PClass *cls)
{
if (Type == EX_NOP || Type == EX_Const || Type == EX_Var)
{
return;
}
else if (Type == EX_Compl || Type == EX_Not || /*Type == EX_Plus || */Type == EX_Minus)
{
if (Children[0]->Type == EX_Const)
{
Value = EvalExpression (this, NULL, cls);
Type = EX_Const;
}
}
else if (Type == EX_Cond)
{
if (Children[0]->Type == EX_Const)
{
bool cond = (Children[0]->Value.Type == VAL_Int) ? (Children[0]->Value.Int != 0) : (Children[0]->Value.Float != 0);
ExpData *data = Children[1]->Children[cond];
delete Children[1]->Children[!cond];
delete Children[1];
delete Children[0];
Type = data->Type;
Value = data->Value;
for (int i = 0; i < 2; i++)
{
Children[i] = data->Children[i];
data->Children[i] = NULL;
}
delete data;
}
}
else if (Type != EX_Random && Type != EX_Sin && Type != EX_Cos && Type != EX_ActionSpecial)
{
if (Children[0]->Type == EX_Const && Children[1]->Type == EX_Const)
{
Value = EvalExpression (this, NULL, NULL);
Type = EX_Const;
delete Children[0]; Children[0] = NULL;
delete Children[1]; Children[1] = NULL;
}
}
}
bool Compare (ExpData *other)
{
if (!other)
return false;
if (Type != other->Type ||
Value.Type != other->Value.Type ||
Value.Float != other->Value.Float ||
RNG != other->RNG)
{
return false;
}
for (int i = 0; i < 2; i++)
{
if (Children[i] && !Children[i]->Compare (other->Children[i]))
return false;
}
return true;
}
ExpOp Type;
ExpVal Value;
ExpData *Children[2];
FRandom *RNG; // only used by random and random2
};
TArray<ExpData *> StateExpressions;
//
// ParseExpression
// [GRB] Parses an expression and stores it into Expression array
//
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionM (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionL (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionK (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionJ (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionI (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionH (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionG (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionF (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionE (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionD (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionC (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionB (FScanner &sc, const PClass *cls);
static ExpData *ParseExpressionA (FScanner &sc, const PClass *cls);
int ParseExpression (FScanner &sc, bool _not, PClass *cls)
2006-04-15 15:00:29 +00:00
{
static bool inited=false;
if (!inited)
{
InitExpressions ();
atterm (ClearExpressions);
inited=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
ExpData *data = ParseExpressionM (sc, cls);
for (unsigned int i = 0; i < StateExpressions.Size (); i++)
{
if (StateExpressions[i]->Compare (data))
{
delete data;
return i;
}
}
return StateExpressions.Push (data);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionM (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionL (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('?'))
{
ExpData *data = new ExpData;
data->Type = EX_Cond;
data->Children[0] = tmp;
ExpData *choices = new ExpData;
data->Children[1] = choices;
choices->Type = EX_Right;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
choices->Children[0] = ParseExpressionM (sc, cls);
sc.MustGetToken(':');
choices->Children[1] = ParseExpressionM (sc, cls);
data->EvalConst (cls);
return data;
}
else
{
return tmp;
}
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionL (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionK (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken(TK_OrOr))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *right = ParseExpressionK (sc, cls);
ExpData *data = new ExpData;
data->Type = EX_LogOr;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionK (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionJ (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken(TK_AndAnd))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *right = ParseExpressionJ (sc, cls);
ExpData *data = new ExpData;
data->Type = EX_LogAnd;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionJ (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionI (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('|'))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *right = ParseExpressionI (sc, cls);
ExpData *data = new ExpData;
data->Type = EX_Or;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionI (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionH (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('^'))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *right = ParseExpressionH (sc, cls);
ExpData *data = new ExpData;
data->Type = EX_Xor;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionH (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionG (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('&'))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *right = ParseExpressionG (sc, cls);
ExpData *data = new ExpData;
data->Type = EX_And;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionG (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionF (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.GetToken() && (sc.TokenType == TK_Eq || sc.TokenType == TK_Neq))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int token = sc.TokenType;
ExpData *right = ParseExpressionF (sc, cls);
ExpData *data = new ExpData;
data->Type = token == TK_Eq? EX_Eq : EX_NE;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.End) sc.UnGet();
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionF (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionE (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.GetToken() && (sc.TokenType == '<' || sc.TokenType == '>' || sc.TokenType == TK_Leq || sc.TokenType == TK_Geq))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int token = sc.TokenType;
ExpData *right = ParseExpressionE (sc, cls);
ExpData *data = new ExpData;
data->Type = token == '<' ? EX_LT : token == '>' ? EX_GT : token == TK_Leq? EX_LE : EX_GE;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.End) sc.UnGet();
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionE (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionD (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.GetToken() && (sc.TokenType == TK_LShift || sc.TokenType == TK_RShift))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int token = sc.TokenType;
ExpData *right = ParseExpressionD (sc, cls);
ExpData *data = new ExpData;
data->Type = token == TK_LShift? EX_LShift : EX_RShift;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.End) sc.UnGet();
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionD (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionC (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.GetToken() && (sc.TokenType == '+' || sc.TokenType == '-'))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int token = sc.TokenType;
ExpData *right = ParseExpressionC (sc, cls);
ExpData *data = new ExpData;
data->Type = token == '+'? EX_Add : EX_Sub;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.End) sc.UnGet();
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionC (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *tmp = ParseExpressionB (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.GetToken() && (sc.TokenType == '*' || sc.TokenType == '/' || sc.TokenType == '%'))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
int token = sc.TokenType;
ExpData *right = ParseExpressionB (sc, cls);
ExpData *data = new ExpData;
data->Type = token == '*'? EX_Mul : token == '/'? EX_Div : EX_Mod;
data->Children[0] = tmp;
data->Children[1] = right;
data->EvalConst (cls);
tmp = data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.End) sc.UnGet();
return tmp;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionB (FScanner &sc, const PClass *cls)
{
ExpData *data = new ExpData;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('~'))
{
data->Type = EX_Compl;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('!'))
{
data->Type = EX_Not;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('-'))
{
data->Type = EX_Minus;
}
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.CheckToken('+');
delete data;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
return ParseExpressionA (sc, cls);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Children[0] = ParseExpressionA (sc, cls);
data->EvalConst (cls);
return data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
static ExpData *ParseExpressionA (FScanner &sc, const PClass *cls)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('('))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
ExpData *data = ParseExpressionM (sc, cls);
sc.MustGetToken(')');
return data;
}
else if (sc.CheckToken(TK_True))
{
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Int;
data->Value.Int = 1;
return data;
}
else if (sc.CheckToken(TK_False))
{
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Int;
data->Value.Int = 0;
return data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken(TK_IntConst))
{
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Int;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Value.Int = sc.Number;
return data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken(TK_FloatConst))
{
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Float;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Value.Float = sc.Float;
return data;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken(TK_Class))
{
// Accept class'SomeClassName'.SomeConstant
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(TK_NameConst);
cls = PClass::FindClass (sc.Name);
if (cls == NULL)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("Unknown class '%s'", sc.String);
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('.');
sc.MustGetToken(TK_Identifier);
PSymbol *sym = cls->Symbols.FindSymbol (sc.String, true);
if (sym != NULL && sym->SymbolType == SYM_Const)
{
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Int;
data->Value.Int = static_cast<PSymbolConst *>(sym)->Value;
return data;
}
else
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("'%s' is not a constant value in class '%s'", sc.String, cls->TypeName.GetChars());
return NULL;
}
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
else if (sc.CheckToken(TK_Identifier))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
switch (FName(sc.String))
{
case NAME_Random:
{
FRandom *rng;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('['))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(TK_Identifier);
rng = FRandom::StaticFindRNG(sc.String);
sc.MustGetToken(']');
}
else
{
rng = &pr_exrandom;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('(');
ExpData *data = new ExpData;
data->Type = EX_Random;
data->RNG = rng;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Children[0] = ParseExpressionM (sc, cls);
sc.MustGetToken(',');
data->Children[1] = ParseExpressionM (sc, cls);
sc.MustGetToken(')');
return data;
}
break;
case NAME_Random2:
{
FRandom *rng;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken('['))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(TK_Identifier);
rng = FRandom::StaticFindRNG(sc.String);
sc.MustGetToken(']');
}
else
{
rng = &pr_exrandom;
}
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('(');
ExpData *data = new ExpData;
data->Type = EX_Random2;
data->RNG = rng;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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.CheckToken(')'))
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Children[0] = ParseExpressionM(sc, cls);
sc.MustGetToken(')');
}
return data;
}
break;
case NAME_Sin:
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('(');
ExpData *data = new ExpData;
data->Type = EX_Sin;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Children[0] = ParseExpressionM (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(')');
return data;
}
break;
case NAME_Cos:
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('(');
ExpData *data = new ExpData;
data->Type = EX_Cos;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Children[0] = ParseExpressionM (sc, cls);
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(')');
return data;
}
break;
default:
{
int specnum, min_args, max_args;
// Check if this is an action special
specnum = P_FindLineSpecial (sc.String, &min_args, &max_args);
if (specnum != 0 && min_args >= 0)
{
int i;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('(');
ExpData *data = new ExpData, **left;
data->Type = EX_ActionSpecial;
data->Value.Int = specnum;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
data->Children[0] = ParseExpressionM (sc, cls);
left = &data->Children[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
for (i = 1; i < 5 && sc.CheckToken(','); ++i)
{
ExpData *right = new ExpData;
right->Type = EX_Right;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
right->Children[0] = ParseExpressionM (sc, cls);
*left = right;
left = &right->Children[1];
}
*left = NULL;
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken(')');
if (i < min_args)
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it 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 ("Not enough arguments to action special");
if (i > max_args)
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.ScriptError ("Too many arguments to action special");
return data;
}
// Check if this is a constant
if (cls != NULL)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
PSymbol *sym = cls->Symbols.FindSymbol (sc.String, true);
if (sym != NULL && sym->SymbolType == SYM_Const)
{
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Int;
data->Value.Int = static_cast<PSymbolConst *>(sym)->Value;
return data;
}
}
// Check if it's a variable we understand
int varid = -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
FName vname = sc.String;
for (size_t i = 0; i < countof(ExpVars); i++)
{
if (vname == ExpVars[i].name)
{
varid = (int)i;
break;
}
}
if (varid == -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 ("Unknown value '%s'", sc.String);
ExpData *data = new ExpData;
data->Type = EX_Var;
data->Value.Type = VAL_Int;
data->Value.Int = varid;
if (ExpVars[varid].array)
{
- Updated lempar.c to v1.31. - Added .txt files to the list of types (wad, zip, and pk3) that can be loaded without listing them after -file. - Fonts that are created by the ACS setfont command to wrap a texture now support animated textures. - FON2 fonts can now use their full palette for CR_UNTRANSLATED when drawn with the hardware 2D path instead of being restricted to the game palette. - Fixed: Toggling vid_vsync would reset the displayed fullscreen gamma to 1 on a Radeon 9000. - Added back the off-by-one palette handling, but in a much more limited scope than before. The skipped entry is assumed to always be at 248, and it is assumed that all Shader Model 1.4 cards suffer from this. That's because all SM1.4 cards are based on variants of the ATI R200 core, and the RV250 in a Radeon 9000 craps up like this. I see no reason to assume that other flavors of the R200 are any different. (Interesting note: With the Radeon 9000, D3DTADDRESS_CLAMP is an invalid address mode when using the debug Direct3D 9 runtime, but it works perfectly fine with the retail Direct3D 9 runtime.) (Insight: The R200 probably uses bytes for all its math inside pixel shaders. That would explain perfectly why I can't use constants greater than 1 with PS1.4 and why it can't do an exact mapping to every entry in the color palette. - Fixed: The software shaded drawer did not work for 2D, because its selected "color"map was replaced with the identitymap before being used. - Fixed: I cannot use Printf to output messages before the framebuffer was completely setup, meaning that Shader Model 1.4 cards could not change resolution. - I have decided to let remap palettes specify variable alpha values for their colors. D3DFB no longer forces them to 255. - Updated re2c to version 0.12.3. - Fixed: A_Wander used threshold as a timer, when it should have used reactiontime. - Fixed: A_CustomRailgun would not fire at all for actors without a target when the aim parameter was disabled. - Made the warp command work in multiplayer, again courtesy of Karate Chris. - Fixed: Trying to spawn a bot while not in a game made for a crashing time. (Patch courtesy of Karate Chris.) - Removed some floating point math from hu_scores.cpp that somebody's GCC gave warnings for (not mine, though). - Fixed: The SBarInfo drawbar command crashed if the sprite image was unavailable. - Fixed: FString::operator=(const char *) did not release its old buffer when being assigned to the null string. - The scanner no longer has an upper limit on the length of strings it accepts, though short strings will be faster than long ones. - Moved all the text scanning functions into a class. Mainly, this means that multiple script scanner states can be stored without being forced to do so recursively. I think I might be taking advantage of that in the near future. Possibly. Maybe. - Removed some potential buffer overflows from the decal parser. - Applied Blzut3's SBARINFO update #9: * Fixed: When using even length values in drawnumber it would cap to a 98 value instead of a 99 as intended. * The SBarInfo parser can now accept negatives for coordinates. This doesn't allow much right now, but later I plan to add better fullscreen hud support in which the negatives will be more useful. This also cleans up the source a bit since all calls for (x, y) coordinates are with the function getCoordinates(). - Added support for stencilling actors. - Added support for non-black colors specified with DTA_ColorOverlay to the software renderer. - Fixed: The inverse, gold, red, and green fixed colormaps each allocated space for 32 different colormaps, even though each only used the first one. - Added two new blending flags to make reverse subtract blending more useful: STYLEF_InvertSource and STYLEF_InvertOverlay. These invert the color that gets blended with the background, since that seems like a good idea for reverse subtraction. They also work with the other two blending operations. - Added subtract and reverse subtract blending operations to the renderer. Since the ERenderStyle enumeration was getting rather unwieldy, I converted it into a new FRenderStyle structure that lets each parameter of the blending equation be set separately. This simplified the set up for the blend quite a bit, and it means a number of new combinations are available by setting the parameters properly. SVN r710 (trunk)
2008-01-25 23:57:44 +00:00
sc.MustGetToken('[');
data->Children[0] = ParseExpressionM (sc, cls);
sc.MustGetToken(']');
}
return data;
}
break;
}
}
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
FString tokname = sc.TokenName(sc.TokenType, sc.String);
sc.ScriptError ("Unexpected token %s", tokname.GetChars());
return NULL;
2006-04-15 15:00:29 +00:00
}
}
//
// EvalExpression
// [GRB] Evaluates previously stored expression
//
bool IsExpressionConst(int id)
{
if (StateExpressions.Size() <= (unsigned int)id) return false;
return StateExpressions[id]->Type == EX_Const;
}
int EvalExpressionI (int id, AActor *self, const PClass *cls)
2006-04-15 15:00:29 +00:00
{
if (StateExpressions.Size() <= (unsigned int)id) return 0;
if (cls == NULL && self != NULL)
{
cls = self->GetClass();
}
ExpVal val = EvalExpression (StateExpressions[id], self, cls);
switch (val.Type)
{
default:
case VAL_Int:
return val.Int;
case VAL_Float:
return (int)val.Float;
}
2006-04-15 15:00:29 +00:00
}
float EvalExpressionF (int id, AActor *self, const PClass *cls)
2006-04-15 15:00:29 +00:00
{
if (StateExpressions.Size() <= (unsigned int)id) return 0.f;
if (cls == NULL && self != NULL)
{
cls = self->GetClass();
}
ExpVal val = EvalExpression (StateExpressions[id], self, cls);
switch (val.Type)
{
default:
case VAL_Int:
return (float)val.Int;
case VAL_Float:
return val.Float;
}
}
fixed_t EvalExpressionFix (int id, AActor *self, const PClass *cls)
{
if (StateExpressions.Size() <= (unsigned int)id) return 0;
if (cls == NULL && self != NULL)
{
cls = self->GetClass();
}
ExpVal val = EvalExpression (StateExpressions[id], self, cls);
switch (val.Type)
{
default:
case VAL_Int:
return val.Int << FRACBITS;
case VAL_Float:
return fixed_t(val.Float*FRACUNIT);
}
}
static ExpVal EvalExpression (ExpData *data, AActor *self, const PClass *cls)
{
ExpVal val;
val.Type = VAL_Int; // Placate GCC
switch (data->Type)
{
case EX_NOP:
assert (data->Type != EX_NOP);
val = data->Value;
break;
case EX_Const:
val = data->Value;
break;
case EX_Var:
if (!self)
{
I_FatalError ("Missing actor data");
}
else
{
int id = 0;
if (ExpVars[data->Value.Int].array)
{
ExpVal idval = EvalExpression (data->Children[0], self, cls);
id = ((idval.Type == VAL_Int) ? idval.Int : (int)idval.Float) % ExpVars[data->Value.Int].array;
}
val = ExpVars[data->Value.Int].get (self, id);
}
break;
case EX_Compl:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
val.Type = VAL_Int;
val.Int = ~((a.Type == VAL_Int) ? a.Int : (int)a.Float);
}
break;
case EX_Not:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
val.Type = VAL_Int;
val.Int = !((a.Type == VAL_Int) ? a.Int : (int)a.Float);
}
break;
case EX_Minus:
{
val = EvalExpression (data->Children[0], self, cls);
if (val.Type == VAL_Int)
val.Int = -val.Int;
else
val.Float = -val.Float;
}
break;
case EX_Mul:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
{
val.Type = VAL_Int;
val.Int = a.Int * b.Int;
}
else
{
val.Type = VAL_Float;
val.Float = a.Int * b.Float;
}
}
else
{
val.Type = VAL_Float;
if (b.Type == VAL_Int)
val.Float = a.Float * b.Int;
else
val.Float = a.Float * b.Float;
}
}
break;
case EX_Div:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
if (b.Int == 0)
{
I_FatalError ("Division by zero");
}
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
{
val.Type = VAL_Int;
val.Int = a.Int / b.Int;
}
else
{
val.Type = VAL_Float;
val.Float = a.Int / b.Float;
}
}
else
{
val.Type = VAL_Float;
if (b.Type == VAL_Int)
val.Float = a.Float / b.Int;
else
val.Float = a.Float / b.Float;
}
}
break;
case EX_Mod:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
if (b.Int == 0)
{
I_FatalError ("Division by zero");
}
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
{
val.Type = VAL_Int;
val.Int = a.Int % b.Int;
}
else
{
val.Type = VAL_Float;
val.Float = fmodf (a.Int, b.Float);
}
}
else
{
val.Type = VAL_Float;
if (b.Type == VAL_Int)
val.Float = fmodf (a.Float, b.Int);
else
val.Float = fmodf (a.Float, b.Float);
}
}
break;
case EX_Add:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
{
val.Type = VAL_Int;
val.Int = a.Int + b.Int;
}
else
{
val.Type = VAL_Float;
val.Float = a.Int + b.Float;
}
}
else
{
val.Type = VAL_Float;
if (b.Type == VAL_Int)
val.Float = a.Float + b.Int;
else
val.Float = a.Float + b.Float;
}
}
break;
case EX_Sub:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
{
val.Type = VAL_Int;
val.Int = a.Int - b.Int;
}
else
{
val.Type = VAL_Float;
val.Float = a.Int - b.Float;
}
}
else
{
val.Type = VAL_Float;
if (b.Type == VAL_Int)
val.Float = a.Float - b.Int;
else
val.Float = a.Float - b.Float;
}
}
break;
case EX_LShift:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int << b.Int;
else
val.Int = a.Int << (int)b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = (int)a.Float << b.Int;
else
val.Int = (int)a.Float << (int)b.Float;
}
}
break;
case EX_RShift:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int >> b.Int;
else
val.Int = a.Int >> (int)b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = (int)a.Float >> b.Int;
else
val.Int = (int)a.Float >> (int)b.Float;
}
}
break;
case EX_LT:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int < b.Int;
else
val.Int = a.Int < b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float < b.Int;
else
val.Int = a.Float < b.Float;
}
}
break;
case EX_GT:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int > b.Int;
else
val.Int = a.Int > b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float > b.Int;
else
val.Int = a.Float > b.Float;
}
}
break;
case EX_LE:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int <= b.Int;
else
val.Int = a.Int <= b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float <= b.Int;
else
val.Int = a.Float <= b.Float;
}
}
break;
case EX_GE:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int >= b.Int;
else
val.Int = a.Int >= b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float >= b.Int;
else
val.Int = a.Float >= b.Float;
}
}
break;
case EX_Eq:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int == b.Int;
else
val.Int = a.Int == b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float == b.Int;
else
val.Int = a.Float == b.Float;
}
}
break;
case EX_NE:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int != b.Int;
else
val.Int = a.Int != b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float != b.Int;
else
val.Int = a.Float != b.Float;
}
}
break;
case EX_And:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int & b.Int;
else
val.Int = a.Int & (int)b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = (int)a.Float & b.Int;
else
val.Int = (int)a.Float & (int)b.Float;
}
}
break;
case EX_Xor:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int ^ b.Int;
else
val.Int = a.Int ^ (int)b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = (int)a.Float ^ b.Int;
else
val.Int = (int)a.Float ^ (int)b.Float;
}
}
break;
case EX_Or:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int | b.Int;
else
val.Int = a.Int | (int)b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = (int)a.Float | b.Int;
else
val.Int = (int)a.Float | (int)b.Float;
}
}
break;
case EX_LogAnd:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int && b.Int;
else
val.Int = a.Int && b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float && b.Int;
else
val.Int = a.Float && b.Float;
}
}
break;
case EX_LogOr:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
val.Type = VAL_Int;
if (a.Type == VAL_Int)
{
if (b.Type == VAL_Int)
val.Int = a.Int || b.Int;
else
val.Int = a.Int || b.Float;
}
else
{
if (b.Type == VAL_Int)
val.Int = a.Float || b.Int;
else
val.Int = a.Float || b.Float;
}
}
break;
case EX_Cond:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
if (a.Type == VAL_Float)
a.Int = (int)a.Float;
val = EvalExpression (data->Children[1]->Children[!!a.Int], self, cls);
}
break;
case EX_Random:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
ExpVal b = EvalExpression (data->Children[1], self, cls);
int min = (a.Type == VAL_Int) ? a.Int : (int)a.Float;
int max = (b.Type == VAL_Int) ? b.Int : (int)b.Float;
val.Type = VAL_Int;
if (max < min)
{
swap (max, min);
}
val.Int = (*data->RNG)(max - min + 1) + min;
}
break;
case EX_Random2:
{
if (data->Children[0] == NULL)
{
val.Type = VAL_Int;
val.Int = data->RNG->Random2();
}
else
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
val.Type = VAL_Int;
val.Int = data->RNG->Random2((a.Type == VAL_Int) ? a.Int : (int)a.Float);
}
}
break;
case EX_Sin:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
angle_t angle = (a.Type == VAL_Int) ? (a.Int * ANGLE_1) : angle_t(a.Float * ANGLE_1);
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (finesine[angle>>ANGLETOFINESHIFT]);
}
break;
case EX_Cos:
{
ExpVal a = EvalExpression (data->Children[0], self, cls);
angle_t angle = (a.Type == VAL_Int) ? (a.Int * ANGLE_1) : angle_t(a.Float * ANGLE_1);
val.Type = VAL_Float;
val.Float = FIXED2FLOAT (finecosine[angle>>ANGLETOFINESHIFT]);
}
break;
case EX_ActionSpecial:
{
int parms[5] = { 0, 0, 0, 0 };
int i = 0;
ExpData *parm = data;
while (parm != NULL && i < 5)
{
ExpVal val = EvalExpression (parm->Children[0], self, cls);
if (val.Type == VAL_Int)
{
parms[i] = val.Int;
}
else
{
parms[i] = (int)val.Float;
}
i++;
parm = parm->Children[1];
}
val.Type = VAL_Int;
val.Int = LineSpecials[data->Value.Int] (NULL, self, false,
parms[0], parms[1], parms[2], parms[3], parms[4]);
}
break;
case EX_Right:
// This should never be a top-level expression.
assert (data->Type != EX_Right);
break;
}
return val;
}
//
// InitExpressions
// [GRB] Set up expression data
//
void InitExpressions ()
{
// StateExpressions[0] always is const 0;
ExpData *data = new ExpData;
data->Type = EX_Const;
data->Value.Type = VAL_Int;
data->Value.Int = 0;
StateExpressions.Push (data);
2006-04-15 15:00:29 +00:00
}
//
// ClearExpressions
// [GRB] Free all expression data
//
void ClearExpressions ()
{
ExpData *data;
while (StateExpressions.Pop (data))
delete data;
}