Difference between revisions of "SDK Known Issues List"

From Valve Developer Community
Jump to: navigation, search
(offsetof errors)
m (offsetof errors)
Line 711: Line 711:
  
 
=== offsetof errors ===
 
=== offsetof errors ===
These are due to [http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16618 a bug in gcc].  It was fixed in 3.4.2, and gcc 3.4.4 is now known to work find and fixes the issue.
+
These are due to [http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16618 a bug in gcc].  It was fixed in 3.4.2, and gcc 3.4.4 is now known to work fine and fixes the issue.
  
 
=== Downcast explicitly ===
 
=== Downcast explicitly ===

Revision as of 20:46, 26 August 2006

Note:A lot of this code is for HL2:MP — yet much is applicable to HL2. If you can't find the said function, don’t fret, it is not in your SDK version.
Note:This page is for the modcode only. For any problems with the SDK tools, see Source SDK Known Issues

Contents

Issues only with SDK 2006-08-04

AI_Activity.cpp assert on startup

Just before the "// Multiplayer" line, add the following:

	ADD_ACTIVITY_TO_SR( ACT_MP_STAND_IDLE_PORTALGUN );
	ADD_ACTIVITY_TO_SR( ACT_MP_RUN_IDLE_PORTALGUN );
	ADD_ACTIVITY_TO_SR( ACT_MP_CROUCH_IDLE_PORTALGUN );
	ADD_ACTIVITY_TO_SR( ACT_MP_WALK_CROUCH_PORTALGUN );
	ADD_ACTIVITY_TO_SR( ACT_MP_GESTURE_RANGE_ATTACK_PORTALGUN );
	ADD_ACTIVITY_TO_SR( ACT_MP_GESTURE_RELOAD_PORTALGUN );
	ADD_ACTIVITY_TO_SR( ACT_MP_JUMP_PORTALGUN );

Will not compile on Linux

This was partially moved to Source SDK Known Issues#Linux_Support

--- src/dlls/hl2_dll/info_darknessmode_lightsource.cpp   2006-08-05 10:58:51.013250000 -0500
+++ src/dlls/hl2_dll/info_darknessmode_lightsource.cpp   2006-08-13 23:36:01.000000000 -0500
@@ -61,7 +61,7 @@

                        if ( g_debug_darkness.GetBool() )
                        {
-                               SetThink( DebugThink );
+                               SetThink( &DebugThink );
                                SetNextThink( gpGlobals->curtime );
                        }
                }
@@ -364,7 +364,7 @@
                {
                        if ( bDebug )
                        {
-                               pLightSource->SetThink( CInfoDarknessLightSource::DebugThink );
+                               pLightSource->SetThink( &CInfoDarknessLightSource::DebugThink );
                                pLightSource->SetNextThink( gpGlobals->curtime );
                        }
                        else

--- src/dlls/episodic/grenade_hopwire.h 2006-08-05 10:58:48.278875000 -0500
+++ src/dlls/episodic/grenade_hopwire.h 2006-08-13 23:09:41.000000000 -0500
@@ -11,7 +11,7 @@
 #endif

 #include "basegrenade_shared.h"
-#include "sprite.h"
+#include "Sprite.h"

 extern ConVar hopwire_trap;


--- src/game_shared/gamestats/ep1_gamestats.cpp  2006-08-05 10:59:06.075750000 -0500
+++ src/game_shared/gamestats/ep1_gamestats.cpp  2006-08-13 16:10:42.000000000 -0500
@@ -5,7 +5,7 @@
 //=============================================================================

 #include "gamestats/ep1_gamestats.h"
-#include "tier1/UtlBuffer.h"
+#include "tier1/utlbuffer.h"

 static int g_nParseVersion = -1;

@@ -194,4 +194,4 @@
                m_nDXLevel = (int)buf.GetShort();
        }
        return bret;
-}
\ No newline at end of file
+}

--- src/game_shared/gamestats/ep1_gamestats.h    2006-08-05 10:59:06.091375000 -0500
+++ src/game_shared/gamestats/ep1_gamestats.h    2006-08-13 16:10:52.000000000 -0500
@@ -11,7 +11,7 @@
 #endif

 class CUtlBuffer;
-#include "tier1/UtlDict.h"
+#include "tier1/utldict.h"

 #define GAMESTATS_FILE_VERSION_OLD             001
 #define GAMESTATS_FILE_VERSION_OLD2            002

That's all the compile bugs I encountered. Granted I was using my mod as a basis, so there might be other compile bugs I didn't see thanks to other fixes in my mod. If anyone else can confirm that's the complete list of fixes for a fresh SDK checkout, please do so.

Will not link fully on Linux

The lack of mathlib results in "undefined symbol: _Z9AngleDiffff" at runtime. There used to be a file called public/mathlib.cpp that exported that symbol, however that file is now complete gone and seemingly has no replacement. (There is a mathlib.lib which is of course useless for non-MSVC.) It's not clear what Valve intended here.

Partial Fix

  • Add back the src/public/mathlib.cpp file that the new SDK deleted.
  • Add back the src/public/mathlib.cpp file that the new SDK deleted.
  • Add the "..\game_shared\choreoevent.cpp" block back to the hl_sdk.vcproj.
  • Add the "..\game_shared\chorescene.cpp" block back to the hl_sdk.vcproj.

Unfortunately then you end up missing a function called void VectorRotate( const Vector &in1, const QAngle &in2, Vector &out ) which never existed in the SDK. I could try to reverse engineer it, but for the moment I'm too tired of dealing with Valve's breakage.

Probably the real solution would be for Valve to ship some .a files containing these symbols, if for some reason they want to hide the source code to mathlib.cpp and friends.

GameInfo.txt issues

This was moved to Source SDK Known Issues#SDK_Launcher_Issues

Animations are jittery

Open cl_dll/c_baseentity.cpp and replace the following code:

if ( !IsSelfAnimating() ) 
{ 
        m_flAnimTime = engine->GetLastTimeStamp();       
}

with this:

if ( !IsSelfAnimating() && !IsNPC() && !IsPlayer() )
	{
		m_flAnimTime = engine->GetLastTimeStamp();	
	}

HL2Gamerules()

When you are running a multiplayer mod, a call to HL2Gamerules(), without checking if the object valid, will probably crash. This happens because the function tries to static_cast g_pGamerules to the class CHalfLife2. However the gamerules of multiplay_gamerules.cpp or hl2dm_gamerules.cpp is used. As a result the casting fails and any calls to it crash. To fix this, you need to add a check to every call to this function.

Bullet impact sounds do not play to clients

In the PlayImpactSound function in cl_dll/fx_impact.cpp, replace:

C_BaseEntity::EmitSound( filter, NULL, pbulletImpactSoundName, pdata->soundhandles.bulletImpact, &vecOrigin );

with:

C_BaseEntity::EmitSound( filter, NULL, pbulletImpactSoundName, &vecOrigin );

This can also be observed as hearing the WateryDeath.Warn soundscript entry played while shooting at models with the "Weapon" surfaceprop when hosting a listen server in a map using level-specific soundscripts.

"Extra App ID set to 211, but no SteamAppID" while debugging

This issue happens when mounting multiple GCFs. The fix is to run the mod via start_mod.bat and let Visual Studio attach to hl2.exe.

Usercmd sends unneeded extra data

The usercmd sends data many times per sec and some of it don't really seem needed.

Impulse commands

Usercmd sends the impulse commands also every time. Unless the impulse commands change VERY often, this is a bandwidth waste. It's better to change this to a normal command.

cl_dll/in_main.cpp

Remove:

static int in_impulse = 0;
void IN_Impulse (void) 
{ 
	in_impulse = atoi( engine->Cmd_Argv(1) ); 
}
// Latch and clear impulse 
cmd->impulse = in_impulse; 
in_impulse = 0;
static ConCommand impulse("impulse", IN_Impulse);

cl_dll/prediction.cpp

Remove:

move->m_nImpulseCommand = ucmd->impulse;

dlls/player.cpp

Remove:

DEFINE_FIELD( m_nImpulse, FIELD_INTEGER ),
ucmd->impulse = 0;
cmd.impulse = ucmd->impulse; 

(The last line is in the file twice)


Change void CBasePlayer::ImpulseCommands( ) to void CBasePlayer::ImpulseCommands( int iImpulse ) . Don't forget the header too.

Add the following code to the top:

// DM: If we are frozen, don't accept impulse-commands
	if ( GetFlags() & FL_FROZEN )
		return;

I guess the comment speaks for itself.

Remove in that function also:

int iImpulse = (int)m_nImpulse; 
m_nImpulse = 0;

Then look for the function CBasePlayer::ClientCommand() and add the following:

else if( stricmp( cmd, "impulse" ) == 0 && engine->Cmd_Argc() == 2 ) 
{ 
	ImpulseCommands( atoi(engine->Cmd_Argv( 1 )) ); 
	return true;
}

Don't forget the HL2DM equivalents.

dlls/player.h

Remove:

int             GetImpulse( void ) const { return m_nImpulse; } 
int                                             m_nImpulse; 

dlls/player_command.cpp

Remove:

move->m_nImpulseCommand         = ucmd->impulse;
if ( ucmd->impulse )
	{
		// Discard impulse commands unless the vehicle allows them.
		// FIXME: UsingStandardWeapons seems like a bad filter for this. The flashlight is an impulse command, for example.
		if ( !pVehicle || player->UsingStandardWeaponsInVehicle() )
		{
			player->m_nImpulse = ucmd->impulse;
		}
	}

game_shared/igamemovement.h

Remove:

int                             m_nImpulseCommand;      // Impulse command issued. 

game_shared/usercmd.cpp

Remove:

if ( to->impulse != from->impulse ) 
{ 
	 	                buf->WriteOneBit( 1 ); 
	 	            buf->WriteUBitLong( to->impulse, 8 ); 
	 	        } 
	 	        else 
	 	        { 
	 	                buf->WriteOneBit( 0 ); 
	 	        } 
if ( buf->ReadOneBit() ) 
	        { 
	 	                move->impulse = buf->ReadUBitLong( 8 ); 
	 	        } 

cl_dll/c_baseplayer.cpp

Remove:

DEFINE_PRED_FIELD( m_nImpulse, FIELD_INTEGER, FTYPEDESC_INSENDTABLE ), 
DEFINE_FIELD( m_nImpulse, FIELD_INTEGER ), 

cl_dll/c_baseplayer.h

Remove:

int                             m_nImpulse;

entitygroundcontact

It also seems that the code lets every client on the server calculate the minimum and maximum height of each animating entity and sends it to the server, while the server doesn't really need it. Each client calculates the same so it's waste of bandwidth.

It's better to test these changes very good because it's possible it may break something.

game_shared/usercmd.cpp

Remove:

#if defined( HL2_CLIENT_DLL )
	if ( to->entitygroundcontact.Count() != 0 )
	{
		buf->WriteOneBit( 1 );
		buf->WriteShort( to->entitygroundcontact.Count() );
		int i;
		for (i = 0; i < to->entitygroundcontact.Count(); i++)
		{
			buf->WriteUBitLong( to->entitygroundcontact[i].entindex, MAX_EDICT_BITS );
			buf->WriteBitCoord( to->entitygroundcontact[i].minheight );
			buf->WriteBitCoord( to->entitygroundcontact[i].maxheight );
		}
		to->entitygroundcontact.RemoveAll();
	}
	else
	{
		buf->WriteOneBit( 0 );
	}
#endif
#if defined( HL2_DLL )
	if ( buf->ReadOneBit() )
	{
		// DM: Fix negative entitygroundcontact crash, from the VDC
		int count = buf->ReadShort();
        if ( count < 0 || count > 140 ) {
            Msg( "Ignoring strange entitygroundcontact count: %d\n", count );
            return; // why can this be less than 0, Valve? change to ReadWord()???
        }
		move->entitygroundcontact.SetCount( count );

		int i;
		for (i = 0; i < move->entitygroundcontact.Count(); i++)
		{
			move->entitygroundcontact[i].entindex = buf->ReadUBitLong( MAX_EDICT_BITS );
			move->entitygroundcontact[i].minheight = buf->ReadBitCoord( );
			move->entitygroundcontact[i].maxheight = buf->ReadBitCoord( );
		}
	}
#endif

game_shared/usercmd.h

Remove:

class CEntityGroundContact
{
public:
	int					entindex;
	float				minheight;
	float				maxheight;
};
#if defined( HL2_DLL ) || defined( HL2_CLIENT_DLL )
		entitygroundcontact.RemoveAll();
#endif
#if defined( HL2_DLL ) || defined( HL2_CLIENT_DLL )
		entitygroundcontact			= src.entitygroundcontact;
#endif
#if defined( HL2_DLL ) || defined( HL2_CLIENT_DLL )
	CUtlVector< CEntityGroundContact > entitygroundcontact;
#endif

dlls/player_command.cpp

Remove:

#if defined (HL2_DLL)
	// pull out backchannel data and move this out

	int i;
	for (i = 0; i < cmd->entitygroundcontact.Count(); i++)
	{
		int entindex =  cmd->entitygroundcontact[i].entindex;
		CBaseEntity *pEntity = CBaseEntity::Instance( engine->PEntityOfEntIndex( entindex) );
		if (pEntity)
		{
			CBaseAnimating *pAnimating = pEntity->GetBaseAnimating();
			if (pAnimating)
			{
				pAnimating->SetIKGroundContactInfo( cmd->entitygroundcontact[i].minheight, cmd->entitygroundcontact[i].maxheight );
			}
		}
	}

#endif

cl_dll/input.h

Remove:
#if defined( HL2_CLIENT_DLL ) 
        CUtlVector< CEntityGroundContact > m_EntityGroundContact; 
#endif

cl_dll/in_main.cpp

Comment out the entire function CInput::AddIKGroundContactInfo


srcds does not inform you that your mod failed to load

You can use the following program to see the message that srcds brokenly fails to display:

// gcc modtest.c -ldl -o modtest
// env LD_LIBRARY_PATH=~/source/bin ./modtest ~/source/xxx/bin/server_i486.so
#include <dlfcn.h>
#include <stdio.h>

int main(int argc, char* argv[]) {
    void* handle = dlopen(argv[1], RTLD_LAZY);
    if (!handle) {
        fprintf(stderr, "error loading .so: %s\n", dlerror());
        return 1;
    }
    void* sym = dlsym(handle, "CreateInterface");
    if (!sym) {
        fprintf(stderr, "error loading symbol - %s\n", dlerror());
        return 1;
    }
    printf("success\n");
    return 0;
}

Resizing game window glitches HUD elements

Fix:

--- mod/src/cl_dll/game_controls/vgui_TeamFortressViewport.cpp  21 Feb 2006 01:58:36 -0000      1.6
+++ mod/src/cl_dll/game_controls/vgui_TeamFortressViewport.cpp  16 Jul 2006 20:52:44 -0000
@@ -152,6 +152,9 @@
        m_pBackGround->SetZPos( -20 ); // send it to the back
        m_pBackGround->SetVisible( false );
        CreateDefaultPanels();
+
+       // reload again... this fixes the ammo display etc
+       ReloadScheme( NULL );
 }

Mod fails to compile with gcc 4.x

Note even if you apply these fixes, gcc 4 still won't work. However both of these fixes seem reasonable on 3.x as well.

Fix: First pass problem is a code bug - not sure if this would've shown up in gcc 3.x if Valve didn't use the horribly evil -w option, but it certainly looks like invalid code to reference a friend function without a declaration.

--- mod/src/dlls/baseentity.h   2005/08/29 00:14:40     1.3
+++ mod/src/dlls/baseentity.h   2006/06/07 02:17:03
@@ -325,6 +325,9 @@
 #define CREATE_PREDICTED_ENTITY( className )   \
        CBaseEntity::CreatePredictedEntityByName( className, __FUNCTION__, __LINE__ );

+void SendProxy_Origin( const SendProp *pProp, const void *pStruct, const void *pData, DVariant *pOut, int iElement, int objectID );
+void SendProxy_Angles( const SendProp *pProp, const void *pStruct, const void *pData, DVariant *pOut, int iElement, int objectID );
+
 //
 // Base Entity.  All entity types derive from this
 //

Looks like the only other issue was this asm change. Apparently gcc did something slightly backwards incompatible in 4.x.

--- src/public/mathlib.cpp      18 Feb 2005 04:45:56 -0000      1.1.1.1
+++ src/public/mathlib.cpp      11 Jun 2006 19:45:34 -0000
@@ -436,7 +436,6 @@
                "movss %%xmm0, %0 \n\t"
                : "=m" (x)
                : "m" (rroot)
-               : "%xmm0"
        );
 #else
 #error

Case changes in player names cause chat truncation

--- mod/src/cl_dll/hl2mp/hl2mp_hud_chat.cpp     2006/02/21 01:58:38     1.3
+++ mod/src/cl_dll/hl2mp/hl2mp_hud_chat.cpp     2006/03/28 03:11:13
@@ -422,8 +422,9 @@
        else
                line->InsertColorChange( Color( g_ColorYellow[0], g_ColorYellow[1], g_ColorYellow[2], 255 ) );

-       char *buf = static_cast<char *>( stackalloc( strlen( pmsg ) + 1  ) );
-       wchar_t *wbuf = static_cast<wchar_t *>( stackalloc( (strlen( pmsg ) + 1 ) * sizeof(wchar_t) ) );
+       char buf[4096];
+       wchar_t wbuf[4096];
+       Assert(strlen(pmsg) < sizeof(buf)); // otherwise message truncation will occur
        if ( buf )
        {
                float *flColor = GetClientColor( iPlayerIndex );
@@ -435,10 +436,10 @@
                buf[ min( iNameLength, MAX_PLAYER_NAME_LENGTH+31) ] = 0;
                line->InsertColorChange( Color( flColor[0], flColor[1], flColor[2], 255 ) );
                line->InsertString( buf );
-               Q_strncpy( buf, pmsg + iNameLength, strlen( pmsg ));
+               Q_strncpy(buf, pmsg + iNameLength, sizeof(buf));
                buf[ strlen( pmsg + iNameLength ) ] = '\0';
                line->InsertColorChange( Color( g_ColorYellow[0], g_ColorYellow[1], g_ColorYellow[2], 255 ) );
-               vgui::localize()->ConvertANSIToUnicode( buf, wbuf, strlen(pmsg)*sizeof(wchar_t));
+               vgui::localize()->ConvertANSIToUnicode( buf, wbuf, sizeof(wbuf));
                line->InsertString( wbuf );
                line->SetVisible( true );
                line->SetNameLength( iNameLength );

--- mod/src/tier1/stringpool.cpp        2005/02/18 04:45:59     1.1.1.1
+++ mod/src/tier1/stringpool.cpp        2006/03/28 03:25:05
@@ -22,7 +22,7 @@

 bool StrLess( const char * const &pszLeft, const char * const &pszRight )
 {
-       return ( Q_stricmp( pszLeft, pszRight) < 0 );
+       return ( Q_strcmp( pszLeft, pszRight) < 0 );
 }

 //-----------------------------------------------------------------------------
--- mod/src/cl_dll/sdk/sdk_hud_chat.cpp	2006/02/21 01:58:38	1.3
+++ mod/src/cl_dll/sdk/sdk_hud_chat.cpp	2006/03/28 03:34:35
@@ -422,8 +422,9 @@
 	else
 		line->InsertColorChange( Color( g_ColorYellow[0], g_ColorYellow[1], g_ColorYellow[2], 255 ) );
 
-	char *buf = static_cast<char *>( stackalloc( strlen( pmsg ) + 1  ) );
-	wchar_t *wbuf = static_cast<wchar_t *>( stackalloc( (strlen( pmsg ) + 1 ) * sizeof(wchar_t) ) );
+	char buf[4096];
+	wchar_t wbuf[4096];
+    assert(strlen(pmsg) < sizeof(buf)); // otherwise message truncation will occur
 	if ( buf )
 	{
 		float *flColor = GetClientColor( iPlayerIndex );
@@ -435,10 +436,10 @@
 		buf[ min( iNameLength, MAX_PLAYER_NAME_LENGTH+31) ] = 0;
 		line->InsertColorChange( Color( flColor[0], flColor[1], flColor[2], 255 ) );
 		line->InsertString( buf );
-		Q_strncpy( buf, pmsg + iNameLength, strlen( pmsg ));
+		Q_strncpy(buf, pmsg + iNameLength, sizeof(buf));
 		buf[ strlen( pmsg + iNameLength ) ] = '\0';
 		line->InsertColorChange( Color( g_ColorYellow[0], g_ColorYellow[1], g_ColorYellow[2], 255 ) );
-		vgui::localize()->ConvertANSIToUnicode( buf, wbuf, strlen(pmsg)*sizeof(wchar_t));
+		vgui::localize()->ConvertANSIToUnicode( buf, wbuf, sizeof(wbuf));
 		line->InsertString( wbuf );
 		line->SetVisible( true );
 		line->SetNameLength( iNameLength );
Note:I've never actually tested the patch to sdk_hud_chat.cpp, but the code is identical to hl2mp_hud_chat.cpp so it should work.

dm_powerhouse lights never slow down

Workaround: Really this should be fixed in the map, but good luck waiting on that to happen. In the mean time here's a code patch for any such mapping errors:

--- mod/src/dlls/physconstraint.cpp     2005/03/26 17:44:28     1.2
+++ mod/src/dlls/physconstraint.cpp     2006/03/06 02:45:47
@@ -727,6 +727,16 @@
                for ( int i = 0; i < 2; i++ )
                {
                        info.pObjects[i]->WorldToLocal( &ballsocket.constraintPosition[i], GetAbsOrigin() );
+            // HACKHACK - the mapper forgot to put in some sane physics damping (dm_powerhouse)
+            float damping, adamping;
+            info.pObjects[i]->GetDamping(&damping, &adamping);
+            if (damping < .2f) {
+                damping = .2f;
+            }
+            if (adamping < .2f) {
+                adamping = .2f;
+            }
+            info.pObjects[i]->SetDamping(&damping, &damping);
                }
                GetBreakParams( ballsocket.constraint, info );
                ballsocket.constraint.torqueLimit = 0;

Wall Boosting Exploit

Hi, everyone would of no doubt noticed that running against a wall(strafing and moving forward works best) makes you run really fast. If anyone has played kreedz climbing yet(I used to code for it) you might of noticed moving against walls is jerky. I think when I was adding some things to the movement I made the server and client disagree on what speed the player is moving at. Has anyone sound why players in source do this? where the boost might be clamped to normal levels? If so I'd love to know, and it would be good to know for bg2 as well, people who play it complain a lot of it as well(just the boost, not client prediction jerks).

Fix Provided by Tony "omega" Sergi

mod/game_shared/gamemovement.cpp Line 1374

void CGameMovement::Accelerate( Vector& wishdir, float wishspeed, float
accel )
{
	int i;
	float addspeed, accelspeed, currentspeed;

	// This gets overridden because some games (CSPort) want to allow dead (observer) players
	// to be able to move around.
	if ( !CanAccelerate() )
		return;

	//omega; WALLSTRAFE FIX

	// See if we are changing direction a bit
	//	currentspeed = mv->m_vecVelocity.Dot(wishdir);

	currentspeed = sqrt( DotProduct(mv->m_vecVelocity, mv->m_vecVelocity) );
	//omega; END WALLSTRAFE FIX

	// Reduce wishspeed by the amount of veer.

bin/tier0_i486.so fails to shut down when using threads

It looks like this would be fixed if Valve would recompile bin/tier0_i486.so with the -fuse-cxa-atexit gcc option. I can't wait for Valve to fix this, so instead:

Workaround: In the mean time what I've done with my mod is add an alarm(10); to the bottom of void CServerGameDLL::DLLShutdown(void) so that if bin/tier0_i486.so fails to exit, it will be killed.

GNU crasher for maps with broken ladders

Fix: Valve's code is invalid but compiles on MSVC due to its D.W.I.M. philosophy. The following patch is required:

--- mod/src/game_shared/func_ladder.cpp 2005/06/11 22:14:32     1.2
+++ mod/src/game_shared/func_ladder.cpp 2006/02/12 19:43:25
@@ -87,7 +87,7 @@
                                m_vecPlayerMountPositionBottom.GetZ(),
                                bottomtrace.m_pEnt
                                        ?
-                                       UTIL_VarArgs( "%s/%s", bottomtrace.m_pEnt->GetClassname(), bottomtrace.m_pEnt->GetEntityName() )
+                                       UTIL_VarArgs( "%s/%s", bottomtrace.m_pEnt->GetClassname(), STRING(bottomtrace.m_pEnt->GetEntityName()) )
                                        :
                                        "NULL" );
                }
@@ -99,7 +99,7 @@
                                m_vecPlayerMountPositionTop.GetZ(),
                                toptrace.m_pEnt
                                        ?
-                                       UTIL_VarArgs( "%s/%s", toptrace.m_pEnt->GetClassname(), toptrace.m_pEnt->GetEntityName() )
+                                       UTIL_VarArgs( "%s/%s", toptrace.m_pEnt->GetClassname(), STRING(toptrace.m_pEnt->GetEntityName()) )
                                        :
                                        "NULL" );
                }

sdkshaders: fix for spaces in path

Issue: Any attempt to compile the SDK shaders for a mod with spaces in the path name results in failure:

(ie: "C:\Program Files\Steam\SteamApps\SourceMods\MODNAME\src\src\sdkshaders")

This is due to the fact that the perl scripts and bat files have not been properly modified to handle spaces in the path.


Fix: Ray "VisualPhoenix" Barbiero has provided the following patches:

Note:These patches have been tested on paths with and without spaces.

updateshaders.pl

--- mod/src/devtools/bin/updateshaders.pl	2005-07-14 09:58:54.000000000 -0500
+++ mod/src/devtools/bin/updateshaders.pl	2005-10-11 23:42:36.000000000 -0500
@@ -105,7 +105,7 @@
 	{
 		$xboxswitch = "-xbox ";
 	}
-	print MAKEFILE "\t$g_SourceDir\\devtools\\bin\\perl.exe $g_SourceDir\\devtools\\bin\\" . $shadertype . "_prep.pl $xboxswitch -shaderoutputdir $shaderoutputdir -source \"$g_SourceDir\" $shadername\n";
+	print MAKEFILE "\t\"$g_SourceDir\\devtools\\bin\\perl.exe\" \"$g_SourceDir\\devtools\\bin\\" . $shadertype . "_prep.pl\" $xboxswitch -shaderoutputdir $shaderoutputdir -source \"$g_SourceDir\" $shadername\n";
 	my $filename;
 	if( $shadertype eq "fxc" )
 	{
@@ -199,14 +199,14 @@
 		# We only generate inc files for fxc and vsh files.
 		if( $g_xbox )
 		{
-			print MAKEFILE " $shadertype" . "tmp_xbox\\" . $shaderbase . "\.inc";
+			print MAKEFILE " \"$shadertype" . "tmp_xbox\\" . $shaderbase . "\.inc\"";
 		}
 		else
 		{
-			print MAKEFILE " $shadertype" . "tmp9\\" . $shaderbase . "\.inc";
+			print MAKEFILE " \"$shadertype" . "tmp9\\" . $shaderbase . "\.inc\"";
 		}
 	}
-	print MAKEFILE " $shaderoutputdir\\$shadertype\\$shaderbase\.vcs";
+	print MAKEFILE " \"$shaderoutputdir\\$shadertype\\$shaderbase\.vcs\"";
 }
 print MAKEFILE "\n\n";
 
@@ -221,11 +221,11 @@
 		# We only generate inc files for fxc and vsh files.
 		if( $g_xbox )
 		{
-			print MAKEFILE "\tdel /f /q $shadertype" . "tmp_xbox\\" . $shaderbase . "\.inc\n";
+			print MAKEFILE "\tdel /f /q \"$shadertype" . "tmp_xbox\\" . $shaderbase . "\.inc\"\n";
 		}
 		else
 		{
-			print MAKEFILE "\tdel /f /q $shadertype" . "tmp9\\" . $shaderbase . "\.inc\n";
+			print MAKEFILE "\tdel /f /q \"$shadertype" . "tmp9\\" . $shaderbase . "\.inc\"\n";
 		}
 	}
 	print MAKEFILE "\tdel /f /q \"$shaderoutputdir\\$shadertype\\$shaderbase\.vcs\"\n";

buildshaders.bat

--- mod/src/materialsystem/stdshaders/buildshaders.bat	2005-07-14 10:00:36.000000000 -0500
+++ mod/src/materialsystem/stdshaders/buildshaders.bat	2005-10-11 23:01:08.000000000 -0500
@@ -36,7 +36,7 @@
 REM ****************
 :set_xbox_args
 set xbox_args=-xbox
-set targetdir=%vproject%\shaders_xbox
+set targetdir="%vproject%\shaders_xbox"
 goto build_shaders
 
 
@@ -45,8 +45,8 @@
 REM ****************
 :set_mod_args
 
-if not exist %sourcesdk%\bin\shadercompile.exe goto NoShaderCompile
-set ChangeToDir=%sourcesdk%\bin
+if not exist "%sourcesdk%\bin\shadercompile.exe" goto NoShaderCompile
+set ChangeToDir="%sourcesdk%\bin"
 
 if /i "%4" NEQ "-source" goto NoSourceDirSpecified
 set SrcDirBase=%~5
@@ -75,7 +75,7 @@
 
 :NoShaderCompile
 echo -
-echo - ERROR: shadercompile.exe doesn't exist in %sourcesdk%\bin
+echo - ERROR: shadercompile.exe doesn't exist in "%sourcesdk%\bin"
 echo -
 goto end
 
@@ -129,7 +129,7 @@
 REM ****************
 REM Execute distributed process on work/build list
 REM ****************
-"%SrcDirBase%\devtools\bin\perl" "%SrcDirBase%\materialsystem\stdshaders\runvmpi.pl" %xbox_args% -changetodir "%ChangeToDir%" %SDKArgs%
+"%SrcDirBase%\devtools\bin\perl" "%SrcDirBase%\materialsystem\stdshaders\runvmpi.pl" %xbox_args% -changetodir %ChangeToDir% %SDKArgs%
 
 REM ****************
 REM Copy the generated files to the output dir.

runvmpi.pl

--- mod/src/materialsystem/stdshaders/runvmpi.pl	2005-07-14 10:00:34.000000000 -0500
+++ mod/src/materialsystem/stdshaders/runvmpi.pl	2005-10-11 22:34:02.000000000 -0500
@@ -41,7 +41,7 @@
 $shaderpath =~ s,/,\\,g;
 chdir $changeToDir;
 
-$cmdToRun = "shadercompile.exe $noMPI $xboxFlag -shaderpath $shaderpath -mpi_workercount 32 -allowdebug $gameFlag";
+$cmdToRun = "shadercompile.exe $noMPI $xboxFlag -shaderpath \"$shaderpath\" -mpi_workercount 32 -allowdebug $gameFlag";
 system $cmdToRun;
 
 # other options..

build_sample_shaders.bat

--- mod/src/sdkshaders/build_sample_shaders.bat	2005-07-14 09:57:02.000000000 -0500
+++ mod/src/sdkshaders/build_sample_shaders.bat	2005-10-11 23:38:18.000000000 -0500
@@ -9,7 +9,7 @@
 
 
 rem **** Call the batch files to build our stuff.
-call ..\materialsystem\stdshaders\buildshaders.bat sdk_shaders -game "%__GameDir%" -source ..
+call ..\materialsystem\stdshaders\buildshaders.bat sdk_shaders -game "%__GameDir%" -source .\..
 goto end


Getting the SDK to work under -Wall -Werror

If you're a professional programmer with any familiarity with GNU then the first thing you probably did was attempt to compile the SDK with -Wall -Werror, and were then disappointed by all the violations in Valve's code out of the box. Here's a list of fixes to make this work. (This list is by no means complete - I fixed many problems before I realized this would be something worth wiki-ing.)

I just noticed Valve uses -fpermissive! *barf*! So, if anyone out there had already noticed that, I now have a work in progress to fix all the scary hacks that -fpermissive would normally error on... coming soon. --Bloodykenny 20:46, 25 Oct 2005 (PDT)

offsetof errors

These are due to a bug in gcc. It was fixed in 3.4.2, and gcc 3.4.4 is now known to work fine and fixes the issue.

Downcast explicitly

One should always downcast explicitly. This squelches useful warnings, and self-documents your intent to downcast.

--- mod/src/public/tier0/fasttimer.h    2005/06/11 22:14:36     1.2
+++ mod/src/public/tier0/fasttimer.h    2005/09/14 00:25:19
@@ -253,7 +253,7 @@
 inline void CCycleCount::Init( float initTimeMsec )
 {
        if ( g_ClockSpeedMillisecondsMultiplier > 0 )
-               m_Int64 = initTimeMsec / g_ClockSpeedMillisecondsMultiplier;
+               m_Int64 = int64(initTimeMsec / g_ClockSpeedMillisecondsMultiplier);
        else
                m_Int64 = 0;
 }

Poor abused templates

  • This code fixes some badly used Memory Templates, and provides compatibility with Unix based systems.
--- mod/src/public/tier1/utlmemory.h    2005/02/18 04:45:58     1.1.1.1
+++ mod/src/public/tier1/utlmemory.h    2005/09/14 00:24:15
@@ -293,7 +293,7 @@
 {
        Assert( num > 0 );

-       if (IsExternallyAllocated())
+    if (CUtlMemory<T>::IsExternallyAllocated())
        {
                // Can't grow a buffer whose memory was externally allocated
                Assert(0);
@@ -349,7 +349,7 @@
        if (m_nAllocationCount >= num)
                return;

-       if (IsExternallyAllocated())
+    if (CUtlMemory<T>::IsExternallyAllocated())
        {
                // Can't grow a buffer whose memory was externally allocated
                Assert(0);
@@ -376,7 +376,7 @@
 template< class T >
 void CUtlMemory<T>::Purge()
 {
-       if (!IsExternallyAllocated())
+    if (!CUtlMemory<T>::IsExternallyAllocated())
        {
                if (m_pMemory)
                {
@@ -495,7 +495,7 @@
 {
        Assert( num > 0 );

-       if (IsExternallyAllocated())
+    if (CUtlMemory<T>::IsExternallyAllocated())
        {
                // Can't grow a buffer whose memory was externally allocated
                Assert(0);
@@ -551,7 +551,7 @@
        if (CUtlMemory<T>::m_nAllocationCount >= num)
                return;

-       if (IsExternallyAllocated())
+    if (CUtlMemory<T>::IsExternallyAllocated())
        {
                // Can't grow a buffer whose memory was externally allocated
                Assert(0);
@@ -579,7 +579,7 @@
 template< class T, int nAlignment >
 void CUtlMemoryAligned<T, nAlignment>::Purge()
 {
-       if (!IsExternallyAllocated())
+    if (!CUtlMemory<T>::IsExternallyAllocated())
        {
                if (m_pMemoryBase)
                {

Missing EOF linefeeds

First you'll need to `cvs rm -f mod/src/dlls/worker_scientist.h` since that file is blank, unused, and will make the script below not run. :)

Then run:

#!/usr/bin/perl -w

use strict;
foreach (`find mod -name *.cpp -or -name *.h`) {
    chomp;
    open(F, "+<", $_) or die "$! on $_";
    seek(F, -1, 2) or die "$! on $_";
    my $c;
    read(F, $c, 1) or die "$! on $_";
    if ($c ne "\n") {
        print "$_\n";
        print(F "\n");
    }
}

Garbage after #endifs

A common mistake in the SDK, obviously trivial to fix, but here for your conveniance:

diff -w -u -r1.3 ai_basenpc.cpp
--- dlls/ai_basenpc.cpp 2005/06/11 22:14:26     1.3
+++ dlls/ai_basenpc.cpp 2005/09/10 18:16:13
@@ -8450,7 +8450,7 @@

        return shotDir;
 }
-#endif HL2_DLL
+#endif // HL2_DLL

 //-----------------------------------------------------------------------------



diff -w -u -r1.1.1.1 npc_BaseZombie.cpp
--- dlls/hl2_dll/npc_BaseZombie.cpp     2005/02/18 04:45:52     1.1.1.1
+++ dlls/hl2_dll/npc_BaseZombie.cpp     2005/09/10 20:57:40
@@ -1582,7 +1582,7 @@

 #ifdef DEBUG_ZOMBIES
                        DevMsg("Wandering\n");
-#endif+
+#endif

                        // Just lost track of our enemy.
                        // Wander around a bit so we don't look like a dingus.



diff -w -u -r1.2 memalloc.h
--- public/tier0/memalloc.h     2005/02/19 18:24:26     1.2
+++ public/tier0/memalloc.h     2005/09/10 05:12:39
@@ -133,6 +133,6 @@
 #define MEM_ALLOC_CREDIT_CLASS()
 #define MEM_ALLOC_CLASSNAME(type) NULL

-#endif !STEAM && NO_MALLOC_OVERRIDE
+#endif // !STEAM && NO_MALLOC_OVERRIDE

 #endif /* TIER0_MEMALLOC_H */




snprintf strangeness

There are several conflicting snprintf defines and other snprintf porting bugs in the SDK, which need to be fixed for Werror compatibility:

diff -w -u -r1.1.1.1 interface.h
--- public/tier1/interface.h    2005/02/18 04:45:58     1.1.1.1
+++ public/tier1/interface.h    2005/09/10 06:09:25
@@ -33,8 +33,6 @@

 #define HMODULE void *
 #define GetProcAddress dlsym
-
-#define _snprintf snprintf
 #endif

 // TODO: move interface.cpp into tier0 library.



diff -w -u -r1.1.1.1 interface.cpp
--- tier1/interface.cpp 2005/02/18 04:45:59     1.1.1.1
+++ tier1/interface.cpp 2005/09/10 21:18:20
@@ -191,7 +191,7 @@
                Q_snprintf( str, sizeof(str), "%s.dll", szAbsoluteModuleName );
                hDLL = LoadLibrary( str );
 #elif _LINUX
-               _snprintf( str, sizeof(str), "%s_i486.so", szAbsoluteModuleName );
+               Q_snprintf( str, sizeof(str), "%s_i486.so", szAbsoluteModuleName );
                 hDLL = dlopen(str, RTLD_NOW);
 #endif

Broken comments

Ending a // comment with \ is just begging for trouble...

diff -w -u -r1.1.1.1 collisionutils.cpp
--- public/collisionutils.cpp   2005/02/18 04:45:56     1.1.1.1
+++ public/collisionutils.cpp   2005/09/10 23:59:09
@@ -2299,7 +2299,7 @@

        return true;
 }
-
+/*
 //--------------------------------------------------------------------------
 // Purpose:
 //
@@ -2314,7 +2314,7 @@
 //    0-----2
 //
 //--------------------------------------------------------------------------
-
+*/
 //-----------------------------------------------------------------------------
 // Purpose: find the minima and maxima of the 3 given values
 //-----------------------------------------------------------------------------

First big pass at -Wall and consistent code

Convert stricmp, _alloca etc to Q_stricmp, stackalloc versions. Fix basetype declarations like 0 to 0.0f when appropriate. Comply with STL min/max. Comply with POSIX/BSD naming inline (and appropriate #ifdefing) in all cases where a cross-platform compatibility utility function does not yet exist.

You'll need to remove the inappropriate -D defines in the Makefile for the str/printf/alloc things. Those are now fixed by using the correct functions in the code to begin with, however I don't have a patch for the Makefile included in the file below.

I have uploaded this file for BloodKenny at -

http://www.nigredostudios.com/temp/patch_hl2_sdk_wall_sanity.txt

http://www.noballs.org/downloads/patch_hl2_sdk_wall_sanity.txt

http://www.ammahls.com/downloads/patch_hl2_sdk_wall_sanity.txt

Right click save as - amckern

Second pass - fix unreliable and error/warning prone min/max/clamp macros with STL calls and normal templatization:

http://aoa.gamemod.net/patch-warning-minmax.txt


Server: Host_Error: SV_PackEntity: SendTable_Encode returned false (ent %d).

Here %d will be replaced by whatever the entity number is that caused the error. No single entity is allowed to write more than 2KB to the network stream. This error happens when an entity overflows that buffer.

Fix: If you need an entity to occasionally transmit more than 2K of data, you will need to split it up into multiple entities, each transmitting part of the data. You should ensure of course that no entity -consistently- transmits 2K of data, as it would bog down your network. Some entities might want to send a large amount on map initialization etc, however. Another possible workaround for some situations (may not always be possible), is to update data a little bit each frame, so that no one frame has a big change.

Server: The client's health variable is pointlessly sent across the wire twice.

Fix: Diff And Patch

  • There is no line 1888 within c_baseplayer.cpp in the stock MP, or SP SDK's - do i just tack the code onto the end (The BaseClass::SetGroundEntity(ground); is also not there) --Amckern 22:46, 5 Nov 2005 (PST)

Ah, sorry. That patch was against an already-modified file, not a fresh SDK. The SetGroundEntity is part of my mod. :) Yes tack it onto the end. After you do so, post the diff back here! --Bloodykenny 09:25, 12 Nov 2005 (PST)

--- mod/src/cl_dll/c_baseplayer.cpp     2005/06/30 05:54:09     1.6
+++ mod/src/cl_dll/c_baseplayer.cpp     2005/07/16 21:33:33
@@ -1888,3 +1888,7 @@
    BaseClass::SetGroundEntity(ground);
}

+
+int C_BasePlayer::GetHealth() const {
+    return g_PR->GetHealth(entindex());
+}



--- mod/src/cl_dll/c_baseplayer.h       2005/07/11 03:28:41     1.9
+++ mod/src/cl_dll/c_baseplayer.h       2005/07/16 21:05:31
@@ -103,7 +103,7 @@

        // Data handlers
        virtual bool    IsPlayer( void ) const { return true; };
-       virtual int             GetHealth() const { return m_iHealth; };
+       virtual int             GetHealth() const;

        // observer mode
        virtual int                     GetObserverMode() const;



--- mod/src/dlls/player.cpp     2005/07/12 07:38:04     1.33
+++ mod/src/dlls/player.cpp     2005/07/22 00:45:48
@@ -58,6 +58,8 @@
 #include "nav_mesh.h"
 #include "env_zoom.h"

+#include "player_resource.h"
+
 #ifdef HL2_DLL
 #include "combine_mine.h"
 #include "weapon_physcannon.h"
@@ -4295,6 +4297,9 @@
        m_lastx = m_lasty = 0;

        m_lastNavArea = NULL;
+
+    // update now so that our new health gets sent with the respawn notice
+    g_pPlayerResource->SetPlayerHealth(entindex(), m_iHealth);

        /// @todo Do this once per round instead of once per player
        if (TheNavMesh)



--- mod/src/dlls/player_resource.cpp    2005/06/19 02:54:04     1.5
+++ mod/src/dlls/player_resource.cpp    2005/07/22 00:44:35
@@ -85,6 +85,13 @@
 }

 //-----------------------------------------------------------------------------
+// Purpose: Sets health value for a particular client (used to set on spawn so the data will be sent synchronously)
+//-----------------------------------------------------------------------------
+void CPlayerResource::SetPlayerHealth(int entindex, int health) {
+    m_iHealth.Set(entindex, health);
+}
+
+//-----------------------------------------------------------------------------
 // Purpose: The Player resource is always transmitted to clients
 //-----------------------------------------------------------------------------
 int CPlayerResource::UpdateTransmitState()



--- mod/src/dlls/player_resource.h      2005/06/18 21:16:29     1.3
+++ mod/src/dlls/player_resource.h      2005/07/22 00:42:57
@@ -26,6 +26,8 @@
        virtual void UpdatePlayerData( void );
        virtual int  UpdateTransmitState(void);

+    void SetPlayerHealth(int entindex, int health);
+
 protected:
        // Data for each player that's propagated to all clients
        // Stored in individual arrays so they can be sent down via datatables

Then you can remove the SendPropInt and RecvPropInt for CBasePlayer::m_iHealth. (However, I don't have a patch for that yet since it's not a backwards-compatible change. Will post a patch once I've done so, if no on else beats me to it.)

Server: The client's team membership is pointlessly sent across the wire twice.

Similar to the above KI, but for another variable. Also this fixes an SDK crasher, but according to Yahn at Valve, Steam will be fixed soon to not do that.

Fix: Diff And Patch

--- mod/src/dlls/team.cpp       2005/06/11 22:14:30     1.2
+++ mod/src/dlls/team.cpp       2006/05/29 18:58:52     1.3
@@ -19,6 +19,7 @@
 //-----------------------------------------------------------------------------
 void SendProxy_PlayerList( const SendProp *pProp, const void *pStruct, const void *pData, DVariant *pOut, int iElement, int objectID )
 {
+    Assert(0);
        CTeam *pTeam = (CTeam*)pData;

        // If this assertion fails, then SendProxyArrayLength_PlayerArray must have failed.
@@ -31,8 +32,11 @@

 int SendProxyArrayLength_PlayerArray( const void *pStruct, int objectID )
 {
-       CTeam *pTeam = (CTeam*)pStruct;
-       return pTeam->m_aPlayers.Count();
+    // The RecvPropArray2 in general was seemingly broken in the May 25th-ish core engine update.
+    // The CTeam in particular was certainly broken at that time, however, thankfully it doesn't
+    // serve any important purpose.
+    // For backwards compatibility with old client, we disable it, rather than remove it entirely.
+    return 0;
 }

Server: weapon_citizensuitcase Assert on startup

Fix: Install copy of the SP SDK, then copy the scripts/ directory over

Server: Some bullet weapons set to do 0 damage.

Fix: Need to edit their AddAmmoType which are strangely set to 0 for some reason.

Server: Some projectile weapons set to do 0 damage.

Fix: Need to edit their source files and set a damage.

Client: --- Missing Vgui material vgui/steam/games/icon_cz

Fix: These files are used in the currently inaccessible Friends IM client. Create 4 vmts in their places of any design to remove the error.

  • materials\VGUI\steam\games\icon_cs.vmt
  • materials\VGUI\steam\games\icon_dod.vmt
  • materials\VGUI\steam\games\icon_hl2mp.vmt
  • materials\VGUI\servers\icon_secure_deny.vmt

Here's a patch to create files for the first three. (I don't see the icon_secure_deny.vmt error, so didn't create anything for it.)

--- materials/VGUI/steam/games/icon_cs.vmt      1969-12-31 18:00:00.000000000 -0600
+++ materials/VGUI/steam/games/icon_cs.vmt      2005-07-28 01:54:17.000000000 -0500
@@ -0,0 +1 @@
+"UnlitGeneric" {}
--- materials/VGUI/steam/games/icon_dod.vmt     1969-12-31 18:00:00.000000000 -0600
+++ materials/VGUI/steam/games/icon_dod.vmt     2005-07-28 01:54:19.000000000 -0500
@@ -0,0 +1 @@
+"UnlitGeneric" {}
--- materials/VGUI/steam/games/icon_hl2mp.vmt   1969-12-31 18:00:00.000000000 -0600
+++ materials/VGUI/steam/games/icon_hl2mp.vmt   2005-07-28 01:54:15.000000000 -0500
@@ -0,0 +1 @@
+"UnlitGeneric" {}

Fix2: Add clear to your rc file in cfg (extract from gcf if non-existant)

Server: The crossbow causes an assertion failure.

Fix: Comment out ai_activity.cpp line 53. This doesn't appear to be something that can be fixed in code?

Server? Client?: RL explosion invisible

Per http://www.chatbear.com/board.plm?a=viewthread&t=970,1107153392,30132&id=786504&b=4991&v=flatold

Fix: in c_te_explosion.cpp

--- mod/src/cl_dll/c_te_explosion.cpp Mon Mar 24 00:46:49 2003
+++ mod/src/cl_dll/c_te_explosion.cpp Mon Mar 24 00:48:38 2003
@@ -241,15 +241,18 @@
const Vector* normal = NULL, unsigned char materialType = 'C' )
{
// Major hack to access singleton object for doing this event (simulate receiving network message)
+ Assert( pos );
+
+ __g_C_TEExplosion.m_vecOrigin = *pos;
__g_C_TEExplosion.m_nModelIndex = modelindex;
__g_C_TEExplosion.m_fScale = scale;
__g_C_TEExplosion.m_nFrameRate = framerate;
__g_C_TEExplosion.m_nFlags = flags;
- __g_C_TEExplosion.m_vecNormal = *normal;
+ __g_C_TEExplosion.m_vecNormal = normal ? *normal : Vector( 0, 0, 1 );
__g_C_TEExplosion.m_chMaterialType = materialType;
__g_C_TEExplosion.m_nRadius = radius;
__g_C_TEExplosion.m_nMagnitude = magnitude;

__g_C_TEExplosion.PostDataUpdate( DATA_UPDATE_CREATED );

}

sv_gravity is not associated with physics gravity

When gravity is changed, the physics gravity isn't updated

Fix: In movevars_shared.cpp:

inline void UpdatePhysicsGravity(float gravity)
{
	if(physenv)
		physenv->SetGravity(Vector(0,0,-gravity));
}
static void GravityChanged_Callback( ConVar *var, const char *pOldString )
{
#ifndef CLIENT_DLL
	UpdatePhysicsGravity(var->GetFloat());
	if(gpGlobals->mapname!=NULL_STRING)
	{
		IGameEvent *event = gameeventmanager->CreateEvent( "gravity_change" );
		if ( event )
		{
			event->SetFloat( "newgravity", var->GetFloat() );
			gameeventmanager->FireEvent( event );
		}
	}
#endif
}
#ifdef CLIENT_DLL
class CGravityChange : public IGameEventListener2, public CAutoGameSystem
{
public:
	bool Init()
	{
		gameeventmanager->AddListener(this,"gravity_change",false);
		return true;
	}
	void FireGameEvent( IGameEvent *event )
	{
		UpdatePhysicsGravity(event->GetFloat("newgravity"));
	}
};
static CGravityChange s_GravityChange;
#endif
ConVar	sv_gravity		( "sv_gravity",DEFAULT_GRAVITY_STRING, FCVAR_NOTIFY | FCVAR_REPLICATED, "World gravity.", GravityChanged_Callback );

In your mod's events file (GameEvents.res/ModEvents.res):

	"gravity_change"
	{
		"newgravity"	"float"
	}

To do: Diff this

Server: util.cpp (531) : Assertion Failed: !"UTIL_GetLocalPlayer"

Occurs when taking damage to yourself from tossing a table straight up into the air.

Fix: The following patch fixes the Assert, and appears to be more functionally correct as well. Validation from Valve that the patch is correct would be appreciated. Diff And Patch

Index: mod/src/dlls/player.h
===================================================================
--- mod/src/dlls/player.h 2005/02/19 22:20:29 1.2
+++ mod/src/dlls/player.h 2005/02/24 00:35:30
@@ -459,6 +459,7 @@
     // mass/size limit set to zero for none
     static bool CanPickupObject( CBaseEntity *pObject, float massLimit, float sizeLimit );
     virtual void PickupObject( CBaseEntity *pObject, bool bLimitMassAndSize = true ) {}
+    virtual bool IsHoldingEntity( CBaseEntity *pEnt );
     virtual void ForceDropOfCarriedPhysObjects( CBaseEntity *pOnlyIfHoldindThis = NULL ) {}
     virtual float GetHeldObjectMass( IPhysicsObject *pHeldObject );

Index: mod/src/dlls/player.cpp
===================================================================
--- mod/src/dlls/player.cpp 2005/02/21 00:05:42 1.3
+++ mod/src/dlls/player.cpp 2005/02/24 00:35:33
@@ -7447,3 +7447,7 @@
     return cmd;
 }

+bool CBasePlayer::IsHoldingEntity( CBaseEntity *pEnt )
+{
+    return PlayerPickupControllerIsHoldingEntity( m_hUseEntity, pEnt );
+}

Index: mod/src/dlls/physics_impact_damage.cpp
===================================================================
--- mod/src/dlls/physics_impact_damage.cpp 2005/02/18 04:45:50 1.1.1.1
+++ mod/src/dlls/physics_impact_damage.cpp 2005/02/24 00:42:24
@@ -417,12 +417,22 @@
     else if ( pEvent->pObjects[index>GetGameFlags() & FVPHYSICS_PLAYER_HELD )
     {
     // if the player is holding the object, use it's real mass (player holding reduced the mass)
-    CBasePlayer *pPlayer = UTIL_GetLocalPlayer();
-    if ( pPlayer )
-    {
+    CBasePlayer* pPlayer = NULL;
+    {for (int i = 1; i <= gpGlobals->maxClients; i++) {
+        CBasePlayer *temp_player = UTIL_PlayerByIndex(i);
+        if (temp_player
+            && temp_player->edict()
+            && temp_player->IsHoldingEntity(pEvent->pEntities[index])
+        ) {
+            pPlayer = temp_player;
+            break;
+        }
+    }}
+    Assert(pPlayer && "object with FVPHYSICS_PLAYER_HELD but no player holding it");
+    if (pPlayer) {
     float mass = pPlayer->GetHeldObjectMass( pEvent->pObjects[index] );
-    if ( mass > 0 )
-    {
+        Assert((mass > 0) && "player was holding object so mass should be non-zero");
+        if (mass > 0) {
             invMass = 1.0f / mass;
         }
     }
Note:Instead of iterating all players, why not consider adding a m_pHolder var and GetHolder and SetHolder functions to CBaseEntity

(You could, though it's unlikely to impact performance much since this is not a frequently called function. Bloodykenny 17:55, 19 Jul 2005 (PDT))

Assert !"UTIL_GetLocalPlayer" when using combine ball

Fix: Diff And Patch

diff -u -r1.1.1.1 prop_combine_ball.cpp 
--- mod/src/dlls/hl2_dll/prop_combine_ball.cpp 2005/02/18 04:45:52 1.1.1.1 
+++ mod/src/dlls/hl2_dll/prop_combine_ball.cpp 2005/02/24 01:34:21 
@@ -788,9 +788,9 @@ 
     pPhysicsObject->GetPosition( &vecPosition, NULL ); 
     pPhysicsObject->GetVelocity( &vecVelocity, NULL ); 

-    CBasePlayer *pPlayer = UTIL_GetLocalPlayer(); 
-    if ( pPlayer ) 
+    if (gpGlobals->maxClients == 1) 
     { 
+    CBasePlayer *pPlayer = UTIL_GetLocalPlayer(); 
     Vector vecDelta; 
     VectorSubtract( pPlayer->GetAbsOrigin(), vecPosition, vecDelta ); 
     VectorNormalize( vecDelta );
Note:This really is a dirty fix because the combine ball always has to face the player because the model was design without a back side


HL2.exe mysteriously segfaults if you forget to put the client.dll and server.dll in your mod's bin directory

Fix: Don't do that.

Update: the behavior here seems to have changed - now HL2 gives an equally mysterious "Server uses different class tables" message. Can anyone else confirm this?

Server/Client: Negative entitygroundcontact crash.

The implications of this SDK bug are not at all clear, but in debug mode you'll get an assert, and in release mode it will presumably segfault. The patch assumes anything negative or really large is bogus and dumps it.

I'm starting to wonder if maybe this is another symptom of the Physics Mayhem bugs. --Bloodykenny 17:24, 4 Nov 2005 (PST)

Fix: Diff And Patch

--- mod/src/game_shared/usercmd.cpp 2005/02/18 04:45:54 1.1.1.1
+++ mod/src/game_shared/usercmd.cpp 2005/03/23 02:24:00
@@ -290,7 +290,12 @@
 #if defined( HL2_DLL )
     if ( buf->ReadOneBit() )
     {
-        move->entitygroundcontact.SetCount( buf->ReadShort() );
+        int count = buf->ReadShort();
+        if ((count < 0) || (count > 140)) {
+            Msg("Ignoring strange entitygroundcontact count: %d\n", count);
+            return; // why can this be less than 0, Valve? change to ReadWord()???
+        }
+    move->entitygroundcontact.SetCount( count );

     int i;
     for (i = 0; i < move->entitygroundcontact.Count(); i++)

Client: Spinning around quickly with the Stun Baton as your active weapon causes the screen to flash white

It's worth noting that recent updates to HL2DM have fixed this. However there has not yet been an SDK code update to show what the fix is. Basically Valve seems to have made the glow fade out completely in between swings.

Fix: Fix unknown

Assert on line 866 of mod/src/game_shared/basecombatweapon_shared.cpp

Fix: I haven't been able to determine exactly why this is an assert. It's just "Assert(0);" with no accompanying explanation, but commenting it out results in no problems it seems. Diff And Patch

--- mod/src/game_shared/basecombatweapon_shared.cpp     2005/02/18 04:45:53     1.1
+++ mod/src/game_shared/basecombatweapon_shared.cpp     2005/03/04 03:45:36     1.2
@@ -863,7 +863,7 @@
        CBaseViewModel *vm = pOwner->GetViewModel( m_nViewModelIndex );
        if ( vm == NULL )
        {
-               Assert( false );
+               //Assert( false );
                return false;
        }

Client: The mod can appear to have frozen on initial startup

Originally posted by MattC here: http://www.chatbear.com/board.plm?a=viewthread&b=4991&t=234,1117553034,6915&s=0&id=868372

Fix: Issue a "progress_enable" command just before "map <mapname>".

engine->ClientCmd("progress_enable");

engine->ClientCmd("map test");

This should fix local servers. For connecting to a remote multiplayer server, the only issue appears to be that you can't switch to another window and then switch back to HL2. Doing so causes HL2 to never render progress info until load is fully complete. That seems to be an HL2 core bug unrelated (though very similar in appearance) to the fix above.

USE_VCR_MODE linker errors

This is because Valve wants to be able to somehow replay all of the things that happen in a mod and get back the exact same output. I'm still confused as to how that would work, or what the value really is, but anyway, if you need things like threading, you'll hit these errors.

My mod needed threading and commenting these out has never caused any issue. Though apparently it means my mod can't use VCR mode.

Fix: Comment out the appropriate lines in protect_things.h, for whatever VCR thing was failing. Note, don't just block out the whole file - the string part at the top is actually useful.


Source Crashes when alt+tab during background map loading

  • When you alt + tab out of the game when the background map is loading, you will end up with a crash. Debug points to a frame count hack that I have no idea how to fix. Someone that knows how to add a curTime event may be able to fix this bug.

Here's a stack trace following a screen change and a crash:

>	engine.dll!0da99893() 	
 	engine.dll!0db000e2() 	
 	engine.dll!0db0046f() 	
 	engine.dll!0dafba15() 	
 	engine.dll!0da9995a() 	
 	engine.dll!0da6247d() 	
 	engine.dll!0db2e05c() 	
 	engine.dll!0d9e3907() 	
 	engine.dll!0da6f816() 	
 	engine.dll!0da7a31c() 	
 	engine.dll!0da7a9b3() 	
 	engine.dll!0da853b5() 	
 	engine.dll!0da854a2() 	
 	engine.dll!0da8553f() 	
 	engine.dll!0db2309c() 	
 	engine.dll!0db22c67() 	
 	MaterialSystem.dll!0d58d0c5() 	
 	engine.dll!0daa50fb() 	
 	engine.dll!0db22eee() 	
 	engine.dll!0db22d24() 	
 	engine.dll!0db22d2f() 	
 	engine.dll!0db22dcc() 	
 	launcher.dll!1000369b() 	
 	launcher.dll!10007f70() 	
 	launcher.dll!10007f70() 	
 	launcher.dll!100057f8() 	

Lighting in Compile (RAD)

There is currently a rather severe lighting bug during RAD. Visible symptoms include: black models while in shadows cast by light_environment; large, unlit sections in "outdoor" type areas.

Fix: Pending.

Using the Sheild Scanner npc_cscanner is not possible unless you name the map with d3_c17 as a prefix

I was expecting the flag for strider scout scanner to be the sheild scanner version of the npc_cscanner, but it seems that the flag is there, but currently in the default hl2 code it does nothing.

So I replaced in the Scanner constructor function the code that compares the first 6 letters of a map name to d3_c17 with

npc_scanner.cpp Line 588

	DevMsg("CNPC_CScanner::Spawn: Invalid spotlight width %.1f (max %.1f).\n", m_flSpotlightGoalWidth, MAX_BEAM_WIDTH );
		m_flSpotlightGoalWidth = MAX_BEAM_WIDTH; 
	}

+if( IsStriderScout() )
+	 {
+		 // Streetwar scanners are claw scanners
+		 m_bIsClawScanner = true;
+	 }
+	 else
+	 {
+		 m_bIsClawScanner = false;
+	 }

Precache();

	if( m_bIsClawScanner )

Complaints about tv_delaymapchange in NON SRCTV Mods

Note:This issue is fixed in the August 2006 SDK update

Issue: Console reports an issue about tv_delaymapchange when loading a mod with no SRCTV Support.

mod/game_shared/multiplay_gamerules.cpp

Undone! you don't have to remove it, instead, add FCVAR_REPLICATED like all the other cvar's. It's doing it because it's a shared ConVar in both dll's and not being replicated. --omega

Instead, change to this:

Line 47
ConVar	mp_timelimit( "mp_timelimit",
					  "0",
					  FCVAR_NOTIFY|FCVAR_REPLICATED,
					  "game time per map in minutes" );

ConVar	tv_delaymapchange( "tv_delaymapchange",
					  "0",
-					  0,
+					  FCVAR_NOTIFY|FCVAR_REPLICATED,
					  "Delays map change until broadcast is complete" );
					  

Physics models show up darker under switchable lights

Switchable lights are not taken into account for visibility of physics models when Source starts a map. So whenever a physics model is mostly lit by switchable lights, it will show up much darker when the map starts.

If the switchable lights are turned off and on again, or if a physics model is moved, then the visibility of this physics model is automatically fixed. But if the map is saved and reloaded, all physics models under switchable lights become darker again.

Note:A switchable light is a light with a targetname, that can be triggered to turn on/off.

Fix: Fix unknown


Sticky Player Collisions

I've been chasing a fix for this for months for Dystopia, when two players collide with each other there's massive prediction errors and to the player it looks like they're jerking back and forth. It can mean real trouble when two players attempt to get through a doorway or something, and it sucks if you're trying to do a good melee system.

Anyways, I found out the cause of this problem is on the server, when it runs a players movement code, it first unlags all the other players on the server, presumably to help predict player vs. player collisions. The problem is, because two players have different pings, there's no authoritative point of impact for the collision, so the jerking around you see is caused by this unlagging. Now down to the fix, which is really simple:

player_command.cpp Line 42

	// Move other players back to history positions based on local player's lag
-	lagcompensation->StartLagCompensation( player, cmd );

player_command.cpp Line 52

	// Restore other players to current positions
-	lagcompensation->FinishLagCompensation( player );

player_command.cpp Line 348

	// Let server invoke any needed impact functions
	moveHelper->ProcessImpacts();

+	// Only do unlag for post think/post frame, which is mainly weapons stuff! - Teddy
+	lagcompensation->StartLagCompensation( player, ucmd );
	RunPostThink( player );
+	lagcompensation->FinishLagCompensation( player );

	FinishCommand( player );

Pretty simple huh? You only need unlag for weapons really, and all the weapon hitscan code is run from the weapon's ItemPostFrame which is what the RunPostThink() runs. Let me know if it works for you! --Teddy 20:05, 12 Jun 2006 (PDT)


Minor/misc code neatenings/fixings

(This section is for issues in the codebase that you've spotted to be potentially dangerous but aren't actually causing obvious bugs/issues... every little improvement to the quality of the codebase helps, right?)

/src/cl_dll/game_controls/spectatorgui.cpp(498): The memset is setting twice as much memory as it should be to 0x0 because sizeof(playerName) already accounts for the fact that the elements are wchar_t type. Fix: Remove the "* sizeof( wchar_t)".


filter_damage_type doesn't work

Open dlls/filters.cpp and replace

return info.GetDamageType() == m_iDamageType;

with

return ( (info.GetDamageType() & m_iDamageType ) ? true : false );

Floating SLAM's

When attaching a SLAM at a movable object, it doesn't explode when the object moves and so give a "floating SLAM".

Fix:

--- mod/src/dlls/hl2_dll/grenade_tripmine.h	18 Feb 2005 04:45:51 -0000	1.1.1.1
+++ mod/src/dlls/hl2_dll/grenade_tripmine.h	22 Jul 2006 22:32:28 -0000
@@ -36,6 +36,8 @@
 	void MakeBeam( void );
 	void KillBeam( void );
 
+    void AttachToEntity(const CBaseEntity* const ent);
+
 public:
 	EHANDLE		m_hOwner;
 
@@ -49,6 +51,10 @@
 	Vector		m_posOwner;
 	Vector		m_angleOwner;
 
+    const CBaseEntity* m_pAttachedObject;
+    Vector m_vecOldPosAttachedObject;
+    QAngle m_vecOldAngAttachedObject;
+
 	DECLARE_DATADESC();
 };
 
--- mod/src/dlls/hl2mp_dll/grenade_tripmine.cpp	18 Feb 2005 04:45:51 -0000	1.1.1.1
+++ mod/src/dlls/hl2mp_dll/grenade_tripmine.cpp	22 Jul 2006 22:32:28 -0000
@@ -91,6 +91,8 @@
 	m_vecEnd = GetAbsOrigin() + m_vecDir * 2048;
 
 	AddEffects( EF_NOSHADOW );
+
+    m_pAttachedObject = NULL;
 }
 
 
@@ -224,7 +226,27 @@
 		return;
 	}
 
-	SetNextThink( gpGlobals->curtime + 0.05f );
+    // Detonate if the parent object moves
+    if (
+        m_pAttachedObject
+        && (
+               !VectorsAreEqual(m_vecOldPosAttachedObject, m_pAttachedObject->GetAbsOrigin(), 1.0f)
+            || !QAnglesAreEqual(m_vecOldAngAttachedObject, m_pAttachedObject->GetAbsAngles(), 1.0f)
+        )
+    ) {
+        m_iHealth = 0;
+        Event_Killed(CTakeDamageInfo((CBaseEntity*)m_hOwner, this, 100, GIB_NORMAL));
+        return;
+    }
+
+	SetNextThink(gpGlobals->curtime + 0.05f);
+}
+
+void CTripmineGrenade::AttachToEntity(const CBaseEntity* const ent) {
+    Assert(m_pAttachedObject == NULL);
+    m_pAttachedObject = ent;
+    m_vecOldPosAttachedObject = ent->GetAbsOrigin();
+    m_vecOldAngAttachedObject = ent->GetAbsAngles();
 }
 
 int CTripmineGrenade::OnTakeDamage_Alive( const CTakeDamageInfo &info )
--- mod/src/dlls/hl2mp_dll/grenade_tripmine.h	18 Feb 2005 04:45:51 -0000	1.1.1.1
+++ mod/src/dlls/hl2mp_dll/grenade_tripmine.h	22 Jul 2006 22:32:28 -0000
@@ -36,6 +36,8 @@
 	void MakeBeam( void );
 	void KillBeam( void );
 
+    void AttachToEntity(const CBaseEntity* const ent);
+
 public:
 	EHANDLE		m_hOwner;
 
@@ -49,6 +51,10 @@
 	Vector		m_posOwner;
 	Vector		m_angleOwner;
 
+    const CBaseEntity* m_pAttachedObject;
+    Vector m_vecOldPosAttachedObject;
+    QAngle m_vecOldAngAttachedObject;
+
 	DECLARE_DATADESC();
 };
 
--- mod/src/game_shared/hl2mp/weapon_slam.cpp	17 Jun 2006 17:51:06 -0000	1.6
+++ mod/src/game_shared/hl2mp/weapon_slam.cpp	22 Jul 2006 22:32:28 -0000
@@ -371,6 +371,7 @@
 			CBaseEntity *pEnt = CBaseEntity::Create( "npc_tripmine", tr.endpos + tr.plane.normal * 3, angles, NULL );
 
 			CTripmineGrenade *pMine = (CTripmineGrenade *)pEnt;
+            pMine->AttachToEntity(pEntity);
 			pMine->m_hOwner = GetOwner();
             pMine->SetDamage(pOwner->get_damage(AOA_SLAM));
 #endif

Other info and links

See also SDK Known Issues List Fixed for the list SDK Known Issues fixed in new SDK revisions.

See also this nice general HL2 bug list on the steam forums. Many of the issues seem imbedded down inside the closed-source side of things, so they are outside the scope of this SDK Known Issues list.

There's also this Patch HL2 petition where users sign their name to beg Valve to fix the aforementioned bugs.

SDK Known Issues specific to memory leaks are on the Memory Leak Fixes page.