1951 lines
52 KiB
C++
1951 lines
52 KiB
C++
// tr_models.c -- model loading and caching
|
|
|
|
#include "tr_local.h"
|
|
|
|
|
|
#include "../qcommon/disablewarnings.h"
|
|
|
|
#pragma warning (push, 3) //go back down to 3 for the stl include
|
|
#include "../qcommon/sstring.h" // #include <string>
|
|
#include <vector>
|
|
#include <map>
|
|
#pragma warning (pop)
|
|
|
|
using namespace std;
|
|
|
|
|
|
#define LL(x) x=LittleLong(x)
|
|
|
|
static qboolean R_LoadMD3 (model_t *mod, int lod, void *buffer, const char *name, qboolean bAlreadyCached );
|
|
static qboolean R_LoadMD4 (model_t *mod, void *buffer, const char *name, qboolean bAlreadyCached );
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
|
|
typedef struct modelHash_s
|
|
{
|
|
char name[MAX_QPATH];
|
|
qhandle_t handle;
|
|
struct modelHash_s *next;
|
|
|
|
}modelHash_t;
|
|
|
|
#define FILE_HASH_SIZE 1024
|
|
static modelHash_t *mhHashTable[FILE_HASH_SIZE];
|
|
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
|
|
|
|
// This stuff looks a bit messy, but it's kept here as black box, and nothing appears in any .H files for other
|
|
// modules to worry about. I may make another module for this sometime.
|
|
//
|
|
typedef pair<int,int> StringOffsetAndShaderIndexDest_t;
|
|
typedef vector <StringOffsetAndShaderIndexDest_t> ShaderRegisterData_t;
|
|
struct CachedEndianedModelBinary_s
|
|
{
|
|
void *pModelDiskImage;
|
|
int iAllocSize; // may be useful for mem-query, but I don't actually need it
|
|
ShaderRegisterData_t ShaderRegisterData;
|
|
int iLastLevelUsedOn;
|
|
int iPAKFileCheckSum; // else -1 if not from PAK
|
|
|
|
|
|
CachedEndianedModelBinary_s()
|
|
{
|
|
pModelDiskImage = 0;
|
|
iAllocSize = 0;
|
|
ShaderRegisterData.clear();
|
|
iLastLevelUsedOn = -1;
|
|
iPAKFileCheckSum = -1;
|
|
}
|
|
};
|
|
typedef struct CachedEndianedModelBinary_s CachedEndianedModelBinary_t;
|
|
typedef map <sstring_t,CachedEndianedModelBinary_t> CachedModels_t;
|
|
CachedModels_t CachedModels; // the important cache item.
|
|
|
|
void RE_RegisterModels_StoreShaderRequest(const char *psModelFileName, const char *psShaderName, int *piShaderIndexPoke)
|
|
{
|
|
char sModelName[MAX_QPATH];
|
|
|
|
Q_strncpyz(sModelName,psModelFileName,sizeof(sModelName));
|
|
Q_strlwr (sModelName);
|
|
|
|
CachedEndianedModelBinary_t &ModelBin = CachedModels[sModelName];
|
|
|
|
if (ModelBin.pModelDiskImage == NULL)
|
|
{
|
|
assert(0); // should never happen, means that we're being called on a model that wasn't loaded
|
|
}
|
|
else
|
|
{
|
|
int iNameOffset = psShaderName - (char *)ModelBin.pModelDiskImage;
|
|
int iPokeOffset = (char*) piShaderIndexPoke - (char *)ModelBin.pModelDiskImage;
|
|
|
|
ModelBin.ShaderRegisterData.push_back( StringOffsetAndShaderIndexDest_t( iNameOffset,iPokeOffset) );
|
|
}
|
|
}
|
|
|
|
|
|
static const byte FakeGLAFile[] =
|
|
{
|
|
0x32, 0x4C, 0x47, 0x41, 0x06, 0x00, 0x00, 0x00, 0x2A, 0x64, 0x65, 0x66, 0x61, 0x75, 0x6C, 0x74,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x3F, 0x01, 0x00, 0x00, 0x00,
|
|
0x14, 0x01, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x18, 0x01, 0x00, 0x00, 0x68, 0x00, 0x00, 0x00,
|
|
0x26, 0x01, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00, 0x4D, 0x6F, 0x64, 0x56, 0x69, 0x65, 0x77, 0x20,
|
|
0x69, 0x6E, 0x74, 0x65, 0x72, 0x6E, 0x61, 0x6C, 0x20, 0x64, 0x65, 0x66, 0x61, 0x75, 0x6C, 0x74,
|
|
0x00, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD,
|
|
0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD,
|
|
0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0xCD, 0x00, 0x00, 0x00, 0x00, 0xFF, 0xFF, 0xFF, 0xFF,
|
|
0x00, 0x00, 0x80, 0x3F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x3F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x3F, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x80, 0x3F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x3F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x3F, 0x00, 0x00, 0x00, 0x00,
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xFD, 0xBF, 0xFE, 0x7F, 0xFE, 0x7F, 0xFE, 0x7F,
|
|
0x00, 0x80, 0x00, 0x80, 0x00, 0x80
|
|
};
|
|
|
|
|
|
// returns qtrue if loaded, and sets the supplied qbool to true if it was from cache (instead of disk)
|
|
// (which we need to know to avoid LittleLong()ing everything again (well, the Mac needs to know anyway)...
|
|
//
|
|
// don't use ri.xxx functions in case running on dedicated...
|
|
//
|
|
qboolean RE_RegisterModels_GetDiskFile( const char *psModelFileName, void **ppvBuffer, qboolean *pqbAlreadyCached)
|
|
{
|
|
char sModelName[MAX_QPATH];
|
|
|
|
Q_strncpyz(sModelName,psModelFileName,sizeof(sModelName));
|
|
Q_strlwr (sModelName);
|
|
|
|
CachedEndianedModelBinary_t &ModelBin = CachedModels[sModelName];
|
|
|
|
if (ModelBin.pModelDiskImage == NULL)
|
|
{
|
|
// didn't have it cached, so try the disk...
|
|
//
|
|
|
|
// special case intercept first...
|
|
//
|
|
if (!strcmp(sDEFAULT_GLA_NAME ".gla" , psModelFileName))
|
|
{
|
|
// return fake params as though it was found on disk...
|
|
//
|
|
void *pvFakeGLAFile = Z_Malloc( sizeof(FakeGLAFile), TAG_FILESYS, qfalse );
|
|
memcpy(pvFakeGLAFile, &FakeGLAFile[0], sizeof(FakeGLAFile));
|
|
*ppvBuffer = pvFakeGLAFile;
|
|
*pqbAlreadyCached = qfalse; // faking it like this should mean that it works fine on the Mac as well
|
|
return qtrue;
|
|
}
|
|
|
|
FS_ReadFile( sModelName, ppvBuffer );
|
|
*pqbAlreadyCached = qfalse;
|
|
qboolean bSuccess = !!(*ppvBuffer)?qtrue:qfalse;
|
|
|
|
if (bSuccess)
|
|
{
|
|
Com_DPrintf( "RE_RegisterModels_GetDiskFile(): Disk-loading \"%s\"\n",psModelFileName);
|
|
}
|
|
|
|
return bSuccess;
|
|
}
|
|
else
|
|
{
|
|
*ppvBuffer = ModelBin.pModelDiskImage;
|
|
*pqbAlreadyCached = qtrue;
|
|
return qtrue;
|
|
}
|
|
}
|
|
|
|
|
|
// if return == true, no further action needed by the caller...
|
|
//
|
|
// don't use ri.xxx functions in case running on dedicated
|
|
//
|
|
extern cvar_t *sv_pure;
|
|
void *RE_RegisterModels_Malloc(int iSize, const char *psModelFileName, qboolean *pqbAlreadyFound, memtag_t eTag)
|
|
{
|
|
char sModelName[MAX_QPATH];
|
|
|
|
Q_strncpyz(sModelName,psModelFileName,sizeof(sModelName));
|
|
Q_strlwr (sModelName);
|
|
|
|
CachedEndianedModelBinary_t &ModelBin = CachedModels[sModelName];
|
|
|
|
if (ModelBin.pModelDiskImage == NULL)
|
|
{
|
|
// ... then this entry has only just been created, ie we need to load it fully...
|
|
//
|
|
void *pMalloc = Z_Malloc(iSize,eTag, qfalse );
|
|
|
|
if (!pMalloc) // not needed anymore, but wtf?
|
|
{
|
|
Com_Error( ERR_FATAL, "RE_RegisterModels_Malloc(): Failed to alloc %d bytes for \"%s\"!", iSize, sModelName);
|
|
}
|
|
|
|
ModelBin.pModelDiskImage = pMalloc;
|
|
ModelBin.iAllocSize = iSize;
|
|
|
|
int iCheckSum;
|
|
if (FS_FileIsInPAK(sModelName, &iCheckSum) == 1)
|
|
{
|
|
ModelBin.iPAKFileCheckSum = iCheckSum; // else ModelBin's constructor will leave it as -1
|
|
}
|
|
|
|
*pqbAlreadyFound = qfalse;
|
|
}
|
|
else
|
|
{
|
|
#ifndef DEDICATED
|
|
// if we already had this model entry, then re-register all the shaders it wanted...
|
|
//
|
|
int iEntries = ModelBin.ShaderRegisterData.size();
|
|
for (int i=0; i<iEntries; i++)
|
|
{
|
|
int iShaderNameOffset = ModelBin.ShaderRegisterData[i].first;
|
|
int iShaderPokeOffset = ModelBin.ShaderRegisterData[i].second;
|
|
|
|
char *psShaderName = &((char*)ModelBin.pModelDiskImage)[iShaderNameOffset];
|
|
int *piShaderPokePtr = (int *) &((char*)ModelBin.pModelDiskImage)[iShaderPokeOffset];
|
|
|
|
shader_t *sh = R_FindShader( psShaderName, lightmapsNone, stylesDefault, qtrue );
|
|
|
|
if ( sh->defaultShader )
|
|
{
|
|
*piShaderPokePtr = 0;
|
|
} else {
|
|
*piShaderPokePtr = sh->index;
|
|
}
|
|
}
|
|
#endif //!DEDICATED
|
|
*pqbAlreadyFound = qtrue; // tell caller not to re-Endian or re-Shader this binary
|
|
}
|
|
|
|
ModelBin.iLastLevelUsedOn = RE_RegisterMedia_GetLevel();
|
|
|
|
return ModelBin.pModelDiskImage;
|
|
}
|
|
|
|
// Unfortunately the dedicated server also hates shader loading. So we need an alternate of this func.
|
|
//
|
|
void *RE_RegisterServerModels_Malloc(int iSize, const char *psModelFileName, qboolean *pqbAlreadyFound, memtag_t eTag)
|
|
{
|
|
char sModelName[MAX_QPATH];
|
|
|
|
Q_strncpyz(sModelName,psModelFileName,sizeof(sModelName));
|
|
Q_strlwr (sModelName);
|
|
|
|
CachedEndianedModelBinary_t &ModelBin = CachedModels[sModelName];
|
|
|
|
if (ModelBin.pModelDiskImage == NULL)
|
|
{
|
|
// ... then this entry has only just been created, ie we need to load it fully...
|
|
//
|
|
void *pMalloc = Z_Malloc(iSize,eTag, qfalse );
|
|
|
|
if (!pMalloc) // not needed anymore, but wtf?
|
|
{
|
|
Com_Error( ERR_FATAL, "RE_RegisterServerModels_Malloc(): Failed to alloc %d bytes for \"%s\"!", iSize, sModelName);
|
|
}
|
|
|
|
ModelBin.pModelDiskImage = pMalloc;
|
|
ModelBin.iAllocSize = iSize;
|
|
|
|
int iCheckSum;
|
|
if (FS_FileIsInPAK(sModelName, &iCheckSum) == 1)
|
|
{
|
|
ModelBin.iPAKFileCheckSum = iCheckSum; // else ModelBin's constructor will leave it as -1
|
|
}
|
|
|
|
*pqbAlreadyFound = qfalse;
|
|
}
|
|
else
|
|
{
|
|
// if we already had this model entry, then re-register all the shaders it wanted...
|
|
//
|
|
/*
|
|
int iEntries = ModelBin.ShaderRegisterData.size();
|
|
for (int i=0; i<iEntries; i++)
|
|
{
|
|
int iShaderNameOffset = ModelBin.ShaderRegisterData[i].first;
|
|
int iShaderPokeOffset = ModelBin.ShaderRegisterData[i].second;
|
|
|
|
char *psShaderName = &((char*)ModelBin.pModelDiskImage)[iShaderNameOffset];
|
|
int *piShaderPokePtr = (int *) &((char*)ModelBin.pModelDiskImage)[iShaderPokeOffset];
|
|
|
|
shader_t *sh = R_FindShader( psShaderName, lightmapsNone, stylesDefault, qtrue );
|
|
|
|
if ( sh->defaultShader )
|
|
{
|
|
*piShaderPokePtr = 0;
|
|
} else {
|
|
*piShaderPokePtr = sh->index;
|
|
}
|
|
}
|
|
*/
|
|
//No. Bad.
|
|
*pqbAlreadyFound = qtrue; // tell caller not to re-Endian or re-Shader this binary
|
|
}
|
|
|
|
ModelBin.iLastLevelUsedOn = RE_RegisterMedia_GetLevel();
|
|
|
|
return ModelBin.pModelDiskImage;
|
|
}
|
|
|
|
// dump any models not being used by this level if we're running low on memory...
|
|
//
|
|
static int GetModelDataAllocSize(void)
|
|
{
|
|
return Z_MemSize( TAG_MODEL_MD3) +
|
|
Z_MemSize( TAG_MODEL_MD4) +
|
|
Z_MemSize( TAG_MODEL_GLM) +
|
|
Z_MemSize( TAG_MODEL_GLA);
|
|
}
|
|
extern cvar_t *r_modelpoolmegs;
|
|
//
|
|
// return qtrue if at least one cached model was freed (which tells z_malloc()-fail recoveryt code to try again)
|
|
//
|
|
extern qboolean gbInsideRegisterModel;
|
|
qboolean RE_RegisterModels_LevelLoadEnd(qboolean bDeleteEverythingNotUsedThisLevel /* = qfalse */)
|
|
{
|
|
qboolean bAtLeastoneModelFreed = qfalse;
|
|
|
|
ri.Printf( PRINT_DEVELOPER, "RE_RegisterModels_LevelLoadEnd():\n");
|
|
|
|
if (gbInsideRegisterModel)
|
|
{
|
|
Com_DPrintf( "(Inside RE_RegisterModel (z_malloc recovery?), exiting...\n");
|
|
}
|
|
else
|
|
{
|
|
int iLoadedModelBytes = GetModelDataAllocSize();
|
|
const int iMaxModelBytes= r_modelpoolmegs->integer * 1024 * 1024;
|
|
|
|
qboolean bEraseOccured = qfalse;
|
|
for (CachedModels_t::iterator itModel = CachedModels.begin(); itModel != CachedModels.end() && ( bDeleteEverythingNotUsedThisLevel || iLoadedModelBytes > iMaxModelBytes ); bEraseOccured?itModel:++itModel)
|
|
{
|
|
bEraseOccured = qfalse;
|
|
|
|
CachedEndianedModelBinary_t &CachedModel = (*itModel).second;
|
|
|
|
qboolean bDeleteThis = qfalse;
|
|
|
|
if (bDeleteEverythingNotUsedThisLevel)
|
|
{
|
|
bDeleteThis = (CachedModel.iLastLevelUsedOn != RE_RegisterMedia_GetLevel()) ? qtrue : qfalse;
|
|
}
|
|
else
|
|
{
|
|
bDeleteThis = (CachedModel.iLastLevelUsedOn < RE_RegisterMedia_GetLevel()) ? qtrue : qfalse;
|
|
}
|
|
|
|
// if it wasn't used on this level, dump it...
|
|
//
|
|
if (bDeleteThis)
|
|
{
|
|
LPCSTR psModelName = (*itModel).first.c_str();
|
|
ri.Printf( PRINT_DEVELOPER, "Dumping \"%s\"", psModelName);
|
|
|
|
#ifdef _DEBUG
|
|
ri.Printf( PRINT_DEVELOPER, ", used on lvl %d\n",CachedModel.iLastLevelUsedOn);
|
|
#endif
|
|
|
|
if (CachedModel.pModelDiskImage) {
|
|
ri.Free(CachedModel.pModelDiskImage);
|
|
//CachedModel.pModelDiskImage = NULL; // REM for reference, erase() call below negates the need for it.
|
|
bAtLeastoneModelFreed = qtrue;
|
|
}
|
|
#ifndef __linux__
|
|
itModel = CachedModels.erase(itModel);
|
|
bEraseOccured = qtrue;
|
|
#else
|
|
// Both MS and Dinkumware got the map::erase wrong
|
|
// The STL has the return type as a void
|
|
CachedModels_t::iterator itTemp;
|
|
itTemp = itModel;
|
|
itModel++;
|
|
CachedModels.erase(itTemp);
|
|
|
|
#endif
|
|
|
|
iLoadedModelBytes = GetModelDataAllocSize();
|
|
}
|
|
}
|
|
}
|
|
|
|
ri.Printf( PRINT_DEVELOPER, "RE_RegisterModels_LevelLoadEnd(): Ok\n");
|
|
|
|
return bAtLeastoneModelFreed;
|
|
}
|
|
|
|
|
|
|
|
// scan through all loaded models and see if their PAK checksums are still valid with the current pure PAK lists,
|
|
// dump any that aren't (so people can't cheat by using models with huge spikes that show through walls etc)
|
|
//
|
|
// (avoid using ri.xxxx stuff here in case running on dedicated)
|
|
//
|
|
static void RE_RegisterModels_DumpNonPure(void)
|
|
{
|
|
Com_DPrintf( "RE_RegisterModels_DumpNonPure():\n");
|
|
|
|
qboolean bEraseOccured = qfalse;
|
|
for (CachedModels_t::iterator itModel = CachedModels.begin(); itModel != CachedModels.end(); bEraseOccured?itModel:++itModel)
|
|
{
|
|
bEraseOccured = qfalse;
|
|
|
|
LPCSTR psModelName = (*itModel).first.c_str();
|
|
CachedEndianedModelBinary_t &CachedModel = (*itModel).second;
|
|
|
|
int iCheckSum = -1;
|
|
int iInPak = FS_FileIsInPAK(psModelName, &iCheckSum);
|
|
|
|
if (iInPak == -1 || iCheckSum != CachedModel.iPAKFileCheckSum)
|
|
{
|
|
if (stricmp(sDEFAULT_GLA_NAME ".gla" , psModelName)) // don't dump "*default.gla", that's program internal anyway
|
|
{
|
|
// either this is not from a PAK, or it's from a non-pure one, so ditch it...
|
|
//
|
|
Com_DPrintf( "Dumping none pure model \"%s\"", psModelName);
|
|
|
|
if (CachedModel.pModelDiskImage) {
|
|
Z_Free(CachedModel.pModelDiskImage);
|
|
//CachedModel.pModelDiskImage = NULL; // REM for reference, erase() call below negates the need for it.
|
|
}
|
|
#ifndef __linux__
|
|
itModel = CachedModels.erase(itModel);
|
|
bEraseOccured = qtrue;
|
|
#else
|
|
// Both MS and Dinkumware got the map::erase wrong
|
|
// The STL has the return type as a void
|
|
CachedModels_t::iterator itTemp;
|
|
itTemp = itModel;
|
|
itModel++;
|
|
CachedModels.erase(itTemp);
|
|
|
|
#endif
|
|
}
|
|
}
|
|
}
|
|
|
|
Com_DPrintf( "RE_RegisterModels_DumpNonPure(): Ok\n");
|
|
}
|
|
|
|
void RE_RegisterModels_Info_f( void )
|
|
{
|
|
int iModels = CachedModels.size();
|
|
int iModel = 0;
|
|
int iTotalBytes = 0;
|
|
for (CachedModels_t::iterator itModel = CachedModels.begin(); itModel != CachedModels.end(); ++itModel,iModel++)
|
|
{
|
|
CachedEndianedModelBinary_t &CachedModel = (*itModel).second;
|
|
|
|
ri.Printf( PRINT_ALL, "%d/%d: \"%s\" (%d bytes)",iModel,iModels,(*itModel).first.c_str(),CachedModel.iAllocSize );
|
|
|
|
#ifdef _DEBUG
|
|
ri.Printf( PRINT_ALL, ", lvl %d\n",CachedModel.iLastLevelUsedOn);
|
|
#endif
|
|
|
|
iTotalBytes += CachedModel.iAllocSize;
|
|
}
|
|
ri.Printf( PRINT_ALL, "%d bytes total (%.2fMB)\n",iTotalBytes, (float)iTotalBytes / 1024.0f / 1024.0f);
|
|
}
|
|
|
|
|
|
// (don't use ri.xxx functions since the renderer may not be running here)...
|
|
//
|
|
static void RE_RegisterModels_DeleteAll(void)
|
|
{
|
|
#ifndef __linux__
|
|
for (CachedModels_t::iterator itModel = CachedModels.begin(); itModel != CachedModels.end(); )
|
|
{
|
|
CachedEndianedModelBinary_t &CachedModel = (*itModel).second;
|
|
|
|
if (CachedModel.pModelDiskImage) {
|
|
Z_Free(CachedModel.pModelDiskImage);
|
|
}
|
|
|
|
itModel = CachedModels.erase(itModel);
|
|
}
|
|
#else
|
|
CachedModels.erase(CachedModels.begin(),CachedModels.end());
|
|
#endif
|
|
}
|
|
|
|
|
|
// do not use ri.xxx functions in here, the renderer may not be running (ie. if on a dedicated server)...
|
|
//
|
|
static int giRegisterMedia_CurrentLevel=0;
|
|
void RE_RegisterMedia_LevelLoadBegin(const char *psMapName, ForceReload_e eForceReload)
|
|
{
|
|
// for development purposes we may want to ditch certain media just before loading a map...
|
|
//
|
|
bool bDeleteModels = eForceReload == eForceReload_MODELS || eForceReload == eForceReload_ALL;
|
|
// bool bDeleteBSP = eForceReload == eForceReload_BSP || eForceReload == eForceReload_ALL;
|
|
|
|
if (bDeleteModels)
|
|
{
|
|
RE_RegisterModels_DeleteAll();
|
|
}
|
|
else
|
|
{
|
|
if (sv_pure->integer)
|
|
{
|
|
RE_RegisterModels_DumpNonPure();
|
|
}
|
|
}
|
|
#ifndef DEDICATED
|
|
// not used in MP codebase...
|
|
//
|
|
// if (bDeleteBSP)
|
|
// {
|
|
// CM_DeleteCachedMap();
|
|
R_Images_DeleteLightMaps(); // always do this now, makes no real load time difference, and lets designers work ok
|
|
// }
|
|
#endif
|
|
|
|
// at some stage I'll probably want to put some special logic here, like not incrementing the level number
|
|
// when going into a map like "brig" or something, so returning to the previous level doesn't require an
|
|
// asset reload etc, but for now...
|
|
//
|
|
// only bump level number if we're not on the same level.
|
|
// Note that this will hide uncached models, which is perhaps a bad thing?...
|
|
//
|
|
static char sPrevMapName[MAX_QPATH]={0};
|
|
if (Q_stricmp( psMapName,sPrevMapName ))
|
|
{
|
|
Q_strncpyz( sPrevMapName, psMapName, sizeof(sPrevMapName) );
|
|
giRegisterMedia_CurrentLevel++;
|
|
}
|
|
}
|
|
|
|
int RE_RegisterMedia_GetLevel(void)
|
|
{
|
|
return giRegisterMedia_CurrentLevel;
|
|
}
|
|
|
|
// this is now only called by the client, so should be ok to dump media...
|
|
//
|
|
extern qboolean SND_RegisterAudio_LevelLoadEnd(qboolean bDeleteEverythingNotUsedThisLevel /* 99% qfalse */);
|
|
extern void S_RestartMusic(void);
|
|
void RE_RegisterMedia_LevelLoadEnd(void)
|
|
{
|
|
RE_RegisterModels_LevelLoadEnd(qfalse);
|
|
#ifndef DEDICATED
|
|
RE_RegisterImages_LevelLoadEnd();
|
|
SND_RegisterAudio_LevelLoadEnd(qfalse);
|
|
// RE_InitDissolve();
|
|
S_RestartMusic();
|
|
#endif
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
** R_GetModelByHandle
|
|
*/
|
|
model_t *R_GetModelByHandle( qhandle_t index ) {
|
|
model_t *mod;
|
|
|
|
// out of range gets the defualt model
|
|
if ( index < 1 || index >= tr.numModels ) {
|
|
return tr.models[0];
|
|
}
|
|
|
|
mod = tr.models[index];
|
|
|
|
return mod;
|
|
}
|
|
|
|
//===============================================================================
|
|
|
|
/*
|
|
** R_AllocModel
|
|
*/
|
|
model_t *R_AllocModel( void ) {
|
|
model_t *mod;
|
|
|
|
if ( tr.numModels == MAX_MOD_KNOWN ) {
|
|
return NULL;
|
|
}
|
|
|
|
mod = (struct model_s *)ri.Hunk_Alloc( sizeof( *tr.models[tr.numModels] ), h_low );
|
|
mod->index = tr.numModels;
|
|
tr.models[tr.numModels] = mod;
|
|
tr.numModels++;
|
|
|
|
return mod;
|
|
}
|
|
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
|
|
/*
|
|
================
|
|
return a hash value for the filename
|
|
================
|
|
*/
|
|
static long generateHashValue( const char *fname, const int size ) {
|
|
int i;
|
|
long hash;
|
|
char letter;
|
|
|
|
hash = 0;
|
|
i = 0;
|
|
while (fname[i] != '\0') {
|
|
letter = tolower(fname[i]);
|
|
if (letter =='.') break; // don't include extension
|
|
if (letter =='\\') letter = '/'; // damn path names
|
|
hash+=(long)(letter)*(i+119);
|
|
i++;
|
|
}
|
|
hash &= (size-1);
|
|
return hash;
|
|
}
|
|
|
|
void RE_InsertModelIntoHash(const char *name, model_t *mod)
|
|
{
|
|
int hash;
|
|
modelHash_t *mh;
|
|
|
|
hash = generateHashValue(name, FILE_HASH_SIZE);
|
|
|
|
// insert this file into the hash table so we can look it up faster later
|
|
mh = (modelHash_t*)ri.Hunk_Alloc( sizeof( modelHash_t ), h_low );
|
|
|
|
mh->next = mhHashTable[hash];
|
|
mh->handle = mod->index;
|
|
strcpy(mh->name, name);
|
|
mhHashTable[hash] = mh;
|
|
}
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
|
|
//rww - Please forgive me for all of the below. Feel free to destroy it and replace it with something better.
|
|
//You obviously can't touch anything relating to shaders or ri. functions here in case a dedicated
|
|
//server is running, which is the entire point of having these seperate functions. If anything major
|
|
//is changed in the non-server-only versions of these functions it would be wise to incorporate it
|
|
//here as well.
|
|
|
|
void GenericInsertModelIntoHash(const char *name, model_t *mod)
|
|
{
|
|
int hash;
|
|
modelHash_t *mh;
|
|
|
|
hash = generateHashValue(name, FILE_HASH_SIZE);
|
|
|
|
// insert this file into the hash table so we can look it up faster later
|
|
mh = (modelHash_t*)Hunk_Alloc( sizeof( modelHash_t ), h_low );
|
|
|
|
mh->next = mhHashTable[hash];
|
|
mh->handle = mod->index;
|
|
strcpy(mh->name, name);
|
|
mhHashTable[hash] = mh;
|
|
}
|
|
|
|
/*
|
|
** GenericAllocModel
|
|
*/
|
|
model_t *GenericAllocModel( void ) {
|
|
model_t *mod;
|
|
|
|
if ( tr.numModels == MAX_MOD_KNOWN ) {
|
|
return NULL;
|
|
}
|
|
|
|
mod = (struct model_s *)Hunk_Alloc( sizeof( *tr.models[tr.numModels] ), h_low );
|
|
mod->index = tr.numModels;
|
|
tr.models[tr.numModels] = mod;
|
|
tr.numModels++;
|
|
|
|
return mod;
|
|
}
|
|
|
|
/*
|
|
=================
|
|
ServerLoadMDXA - load a Ghoul 2 animation file
|
|
=================
|
|
*/
|
|
qboolean ServerLoadMDXA( model_t *mod, void *buffer, const char *mod_name, qboolean bAlreadyCached ) {
|
|
|
|
mdxaHeader_t *pinmodel, *mdxa;
|
|
int version;
|
|
int size;
|
|
|
|
#ifndef _M_IX86
|
|
int j, k, i;
|
|
int frameSize;
|
|
mdxaFrame_t *cframe;
|
|
mdxaSkel_t *boneInfo;
|
|
#endif
|
|
|
|
pinmodel = (mdxaHeader_t *)buffer;
|
|
//
|
|
// read some fields from the binary, but only LittleLong() them when we know this wasn't an already-cached model...
|
|
//
|
|
version = (pinmodel->version);
|
|
size = (pinmodel->ofsEnd);
|
|
|
|
if (!bAlreadyCached)
|
|
{
|
|
version = LittleLong(version);
|
|
size = LittleLong(size);
|
|
}
|
|
|
|
if (version != MDXA_VERSION) {
|
|
return qfalse;
|
|
}
|
|
|
|
mod->type = MOD_MDXA;
|
|
mod->dataSize += size;
|
|
|
|
qboolean bAlreadyFound = qfalse;
|
|
mdxa = mod->mdxa = (mdxaHeader_t*) //ri.Hunk_Alloc( size );
|
|
RE_RegisterServerModels_Malloc(size, mod_name, &bAlreadyFound, TAG_MODEL_GLA);
|
|
|
|
assert(bAlreadyCached == bAlreadyFound); // I should probably eliminate 'bAlreadyFound', but wtf?
|
|
|
|
if (!bAlreadyFound)
|
|
{
|
|
memcpy( mdxa, buffer, size );
|
|
|
|
LL(mdxa->ident);
|
|
LL(mdxa->version);
|
|
LL(mdxa->numFrames);
|
|
LL(mdxa->numBones);
|
|
LL(mdxa->ofsFrames);
|
|
LL(mdxa->ofsEnd);
|
|
}
|
|
|
|
if ( mdxa->numFrames < 1 ) {
|
|
return qfalse;
|
|
}
|
|
|
|
if (bAlreadyFound)
|
|
{
|
|
return qtrue; // All done, stop here, do not LittleLong() etc. Do not pass go...
|
|
}
|
|
|
|
#ifndef _M_IX86
|
|
|
|
//
|
|
// optimisation, we don't bother doing this for standard intel case since our data's already in that format...
|
|
//
|
|
|
|
// swap all the skeletal info
|
|
boneInfo = (mdxaSkel_t *)( (byte *)mdxa + mdxa->ofsSkel);
|
|
for ( i = 0 ; i < mdxa->numBones ; i++)
|
|
{
|
|
LL(boneInfo->numChildren);
|
|
LL(boneInfo->parent);
|
|
for (k=0; k<boneInfo->numChildren; k++)
|
|
{
|
|
LL(boneInfo->children[k]);
|
|
}
|
|
|
|
// get next bone
|
|
boneInfo += (int)( &((mdxaSkel_t *)0)->children[ boneInfo->numChildren ] );
|
|
}
|
|
|
|
|
|
// swap all the frames
|
|
frameSize = (int)( &((mdxaFrame_t *)0)->bones[ mdxa->numBones ] );
|
|
for ( i = 0 ; i < mdxa->numFrames ; i++)
|
|
{
|
|
cframe = (mdxaFrame_t *) ( (byte *)mdxa + mdxa->ofsFrames + i * frameSize );
|
|
cframe->radius = LittleFloat( cframe->radius );
|
|
for ( j = 0 ; j < 3 ; j++ )
|
|
{
|
|
cframe->bounds[0][j] = LittleFloat( cframe->bounds[0][j] );
|
|
cframe->bounds[1][j] = LittleFloat( cframe->bounds[1][j] );
|
|
cframe->localOrigin[j] = LittleFloat( cframe->localOrigin[j] );
|
|
}
|
|
for ( j = 0 ; j < mdxa->numBones * sizeof( mdxaBone_t ) / 2 ; j++ )
|
|
{
|
|
((short *)cframe->bones)[j] = LittleShort( ((short *)cframe->bones)[j] );
|
|
}
|
|
}
|
|
#endif
|
|
return qtrue;
|
|
}
|
|
|
|
/*
|
|
=================
|
|
ServerLoadMDXM - load a Ghoul 2 Mesh file
|
|
=================
|
|
*/
|
|
qboolean ServerLoadMDXM( model_t *mod, void *buffer, const char *mod_name, qboolean bAlreadyCached ) {
|
|
int i,l, j;
|
|
mdxmHeader_t *pinmodel, *mdxm;
|
|
mdxmLOD_t *lod;
|
|
mdxmSurface_t *surf;
|
|
int version;
|
|
int size;
|
|
shader_t *sh;
|
|
mdxmSurfHierarchy_t *surfInfo;
|
|
|
|
#ifndef _M_IX86
|
|
int k;
|
|
int frameSize;
|
|
mdxmTag_t *tag;
|
|
mdxmTriangle_t *tri;
|
|
mdxmVertex_t *v;
|
|
mdxmFrame_t *cframe;
|
|
int *boneRef;
|
|
#endif
|
|
|
|
pinmodel= (mdxmHeader_t *)buffer;
|
|
//
|
|
// read some fields from the binary, but only LittleLong() them when we know this wasn't an already-cached model...
|
|
//
|
|
version = (pinmodel->version);
|
|
size = (pinmodel->ofsEnd);
|
|
|
|
if (!bAlreadyCached)
|
|
{
|
|
version = LittleLong(version);
|
|
size = LittleLong(size);
|
|
}
|
|
|
|
if (version != MDXM_VERSION) {
|
|
return qfalse;
|
|
}
|
|
|
|
mod->type = MOD_MDXM;
|
|
mod->dataSize += size;
|
|
|
|
qboolean bAlreadyFound = qfalse;
|
|
mdxm = mod->mdxm = (mdxmHeader_t*) //ri.Hunk_Alloc( size );
|
|
RE_RegisterServerModels_Malloc(size, mod_name, &bAlreadyFound, TAG_MODEL_GLM);
|
|
|
|
assert(bAlreadyCached == bAlreadyFound); // I should probably eliminate 'bAlreadyFound', but wtf?
|
|
|
|
if (!bAlreadyFound)
|
|
{
|
|
memcpy( mdxm, buffer, size );
|
|
|
|
LL(mdxm->ident);
|
|
LL(mdxm->version);
|
|
LL(mdxm->numLODs);
|
|
LL(mdxm->ofsLODs);
|
|
LL(mdxm->numSurfaces);
|
|
LL(mdxm->ofsSurfHierarchy);
|
|
LL(mdxm->ofsEnd);
|
|
}
|
|
|
|
// first up, go load in the animation file we need that has the skeletal animation info for this model
|
|
mdxm->animIndex = RE_RegisterServerModel(va ("%s.gla",mdxm->animName));
|
|
if (!mdxm->animIndex)
|
|
{
|
|
return qfalse;
|
|
}
|
|
|
|
mod->numLods = mdxm->numLODs -1 ; //copy this up to the model for ease of use - it wil get inced after this.
|
|
|
|
if (bAlreadyFound)
|
|
{
|
|
return qtrue; // All done. Stop, go no further, do not LittleLong(), do not pass Go...
|
|
}
|
|
|
|
surfInfo = (mdxmSurfHierarchy_t *)( (byte *)mdxm + mdxm->ofsSurfHierarchy);
|
|
for ( i = 0 ; i < mdxm->numSurfaces ; i++)
|
|
{
|
|
LL(surfInfo->numChildren);
|
|
LL(surfInfo->parentIndex);
|
|
|
|
// do all the children indexs
|
|
for (j=0; j<surfInfo->numChildren; j++)
|
|
{
|
|
LL(surfInfo->childIndexes[j]);
|
|
}
|
|
|
|
// We will not be using shaders on the server.
|
|
sh = 0;
|
|
// insert it in the surface list
|
|
|
|
surfInfo->shaderIndex = 0;
|
|
|
|
RE_RegisterModels_StoreShaderRequest(mod_name, &surfInfo->shader[0], &surfInfo->shaderIndex);
|
|
|
|
// find the next surface
|
|
surfInfo = (mdxmSurfHierarchy_t *)( (byte *)surfInfo + (int)( &((mdxmSurfHierarchy_t *)0)->childIndexes[ surfInfo->numChildren ] ));
|
|
}
|
|
|
|
// swap all the LOD's (we need to do the middle part of this even for intel, because of shader reg and err-check)
|
|
lod = (mdxmLOD_t *) ( (byte *)mdxm + mdxm->ofsLODs );
|
|
for ( l = 0 ; l < mdxm->numLODs ; l++)
|
|
{
|
|
int triCount = 0;
|
|
|
|
LL(lod->ofsEnd);
|
|
// swap all the surfaces
|
|
surf = (mdxmSurface_t *) ( (byte *)lod + sizeof (mdxmLOD_t) + (mdxm->numSurfaces * sizeof(mdxmLODSurfOffset_t)) );
|
|
for ( i = 0 ; i < mdxm->numSurfaces ; i++)
|
|
{
|
|
LL(surf->numTriangles);
|
|
LL(surf->ofsTriangles);
|
|
LL(surf->numVerts);
|
|
LL(surf->ofsVerts);
|
|
LL(surf->ofsEnd);
|
|
LL(surf->ofsHeader);
|
|
LL(surf->numBoneReferences);
|
|
LL(surf->ofsBoneReferences);
|
|
// LL(surf->maxVertBoneWeights);
|
|
|
|
triCount += surf->numTriangles;
|
|
|
|
if ( surf->numVerts > SHADER_MAX_VERTEXES ) {
|
|
return qfalse;
|
|
}
|
|
if ( surf->numTriangles*3 > SHADER_MAX_INDEXES ) {
|
|
return qfalse;
|
|
}
|
|
|
|
// change to surface identifier
|
|
surf->ident = SF_MDX;
|
|
|
|
// register the shaders
|
|
#ifndef _M_IX86
|
|
//
|
|
// optimisation, we don't bother doing this for standard intel case since our data's already in that format...
|
|
//
|
|
// FIXME - is this correct?
|
|
// do all the bone reference data
|
|
boneRef = (int *) ( (byte *)surf + surf->ofsBoneReferences );
|
|
for ( j = 0 ; j < surf->numBoneReferences ; j++ )
|
|
{
|
|
LL(boneRef[j]);
|
|
}
|
|
|
|
|
|
// swap all the triangles
|
|
tri = (mdxmTriangle_t *) ( (byte *)surf + surf->ofsTriangles );
|
|
for ( j = 0 ; j < surf->numTriangles ; j++, tri++ )
|
|
{
|
|
LL(tri->indexes[0]);
|
|
LL(tri->indexes[1]);
|
|
LL(tri->indexes[2]);
|
|
}
|
|
|
|
// swap all the vertexes
|
|
v = (mdxmVertex_t *) ( (byte *)surf + surf->ofsVerts );
|
|
for ( j = 0 ; j < surf->numVerts ; j++ )
|
|
{
|
|
v->normal[0] = LittleFloat( v->normal[0] );
|
|
v->normal[1] = LittleFloat( v->normal[1] );
|
|
v->normal[2] = LittleFloat( v->normal[2] );
|
|
|
|
v->texCoords[0] = LittleFloat( v->texCoords[0] );
|
|
v->texCoords[1] = LittleFloat( v->texCoords[1] );
|
|
|
|
v->numWeights = LittleLong( v->numWeights );
|
|
v->offset[0] = LittleFloat( v->offset[0] );
|
|
v->offset[1] = LittleFloat( v->offset[1] );
|
|
v->offset[2] = LittleFloat( v->offset[2] );
|
|
|
|
for ( k = 0 ; k < /*v->numWeights*/surf->maxVertBoneWeights ; k++ )
|
|
{
|
|
v->weights[k].boneIndex = LittleLong( v->weights[k].boneIndex );
|
|
v->weights[k].boneWeight = LittleFloat( v->weights[k].boneWeight );
|
|
}
|
|
v = (mdxmVertex_t *)&v->weights[/*v->numWeights*/surf->maxVertBoneWeights];
|
|
}
|
|
#endif
|
|
|
|
// find the next surface
|
|
surf = (mdxmSurface_t *)( (byte *)surf + surf->ofsEnd );
|
|
}
|
|
|
|
// find the next LOD
|
|
lod = (mdxmLOD_t *)( (byte *)lod + lod->ofsEnd );
|
|
}
|
|
|
|
return qtrue;
|
|
}
|
|
|
|
/*
|
|
====================
|
|
RE_RegisterServerModel
|
|
|
|
Same as RE_RegisterModel, except used by the server to handle ghoul2 instance models.
|
|
====================
|
|
*/
|
|
qhandle_t RE_RegisterServerModel( const char *name ) {
|
|
model_t *mod;
|
|
unsigned *buf;
|
|
int lod;
|
|
int ident;
|
|
qboolean loaded;
|
|
// qhandle_t hModel;
|
|
int numLoaded;
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
int hash;
|
|
modelHash_t *mh;
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
|
|
if (!r_noServerGhoul2)
|
|
{ //keep it from choking when it gets to these checks in the g2 code. Registering all r_ cvars for the server would be a Bad Thing though.
|
|
r_noServerGhoul2 = Cvar_Get( "r_noserverghoul2", "0", 0);
|
|
}
|
|
|
|
if ( !name || !name[0] ) {
|
|
return 0;
|
|
}
|
|
|
|
if ( strlen( name ) >= MAX_QPATH ) {
|
|
return 0;
|
|
}
|
|
|
|
hash = generateHashValue(name, FILE_HASH_SIZE);
|
|
|
|
//
|
|
// see if the model is already loaded
|
|
//
|
|
for (mh=mhHashTable[hash]; mh; mh=mh->next) {
|
|
if (Q_stricmp(mh->name, name) == 0) {
|
|
return mh->handle;
|
|
}
|
|
}
|
|
|
|
if ( ( mod = GenericAllocModel() ) == NULL ) {
|
|
return 0;
|
|
}
|
|
|
|
// only set the name after the model has been successfully loaded
|
|
Q_strncpyz( mod->name, name, sizeof( mod->name ) );
|
|
|
|
#ifndef DEDICATED
|
|
// make sure the render thread is stopped
|
|
R_SyncRenderThread();
|
|
#endif
|
|
|
|
int iLODStart = 0;
|
|
if (strstr (name, ".md3")) {
|
|
iLODStart = MD3_MAX_LODS-1; // this loads the md3s in reverse so they can be biased
|
|
}
|
|
mod->numLods = 0;
|
|
|
|
//
|
|
// load the files
|
|
//
|
|
numLoaded = 0;
|
|
|
|
for ( lod = iLODStart; lod >= 0 ; lod-- ) {
|
|
char filename[1024];
|
|
|
|
strcpy( filename, name );
|
|
|
|
if ( lod != 0 ) {
|
|
char namebuf[80];
|
|
|
|
if ( strrchr( filename, '.' ) ) {
|
|
*strrchr( filename, '.' ) = 0;
|
|
}
|
|
sprintf( namebuf, "_%d.md3", lod );
|
|
strcat( filename, namebuf );
|
|
}
|
|
|
|
qboolean bAlreadyCached = qfalse;
|
|
if (!RE_RegisterModels_GetDiskFile(filename, (void **)&buf, &bAlreadyCached))
|
|
{
|
|
continue;
|
|
}
|
|
|
|
//loadmodel = mod; // this seems to be fairly pointless
|
|
|
|
// important that from now on we pass 'filename' instead of 'name' to all model load functions,
|
|
// because 'filename' accounts for any LOD mangling etc so guarantees unique lookups for yet more
|
|
// internal caching...
|
|
//
|
|
ident = *(unsigned *)buf;
|
|
if (!bAlreadyCached)
|
|
{
|
|
ident = LittleLong(ident);
|
|
}
|
|
|
|
switch (ident)
|
|
{ //if you're trying to register anything else as a model type on the server, you are out of luck
|
|
|
|
case MDXA_IDENT:
|
|
loaded = ServerLoadMDXA( mod, buf, filename, bAlreadyCached );
|
|
break;
|
|
case MDXM_IDENT:
|
|
loaded = ServerLoadMDXM( mod, buf, filename, bAlreadyCached );
|
|
break;
|
|
default:
|
|
goto fail;
|
|
}
|
|
|
|
if (!bAlreadyCached){ // important to check!!
|
|
FS_FreeFile (buf);
|
|
}
|
|
|
|
if ( !loaded ) {
|
|
if ( lod == 0 ) {
|
|
goto fail;
|
|
} else {
|
|
break;
|
|
}
|
|
} else {
|
|
mod->numLods++;
|
|
numLoaded++;
|
|
}
|
|
}
|
|
|
|
if ( numLoaded ) {
|
|
// duplicate into higher lod spots that weren't
|
|
// loaded, in case the user changes r_lodbias on the fly
|
|
for ( lod-- ; lod >= 0 ; lod-- ) {
|
|
mod->numLods++;
|
|
mod->md3[lod] = mod->md3[lod+1];
|
|
}
|
|
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
|
|
GenericInsertModelIntoHash(name, mod);
|
|
return mod->index;
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
}
|
|
|
|
fail:
|
|
// we still keep the model_t around, so if the model name is asked for
|
|
// again, we won't bother scanning the filesystem
|
|
mod->type = MOD_BAD;
|
|
return 0;
|
|
}
|
|
|
|
|
|
/*
|
|
====================
|
|
RE_RegisterModel
|
|
|
|
Loads in a model for the given name
|
|
|
|
Zero will be returned if the model fails to load.
|
|
An entry will be retained for failed models as an
|
|
optimization to prevent disk rescanning if they are
|
|
asked for again.
|
|
====================
|
|
*/
|
|
static qhandle_t RE_RegisterModel_Actual( const char *name ) {
|
|
model_t *mod;
|
|
unsigned *buf;
|
|
int lod;
|
|
int ident;
|
|
qboolean loaded;
|
|
// qhandle_t hModel;
|
|
int numLoaded;
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
int hash;
|
|
modelHash_t *mh;
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
|
|
if ( !name || !name[0] ) {
|
|
ri.Printf( PRINT_ALL, "RE_RegisterModel: NULL name\n" );
|
|
return 0;
|
|
}
|
|
|
|
if ( strlen( name ) >= MAX_QPATH ) {
|
|
ri.Printf( PRINT_DEVELOPER, "Model name exceeds MAX_QPATH\n" );
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
// if (!tr.registered) {
|
|
// ri.Printf( PRINT_WARNING, "RE_RegisterModel (%s) called before ready!\n",name );
|
|
// return 0;
|
|
// }
|
|
//
|
|
// search the currently loaded models
|
|
//
|
|
hash = generateHashValue(name, FILE_HASH_SIZE);
|
|
|
|
//
|
|
// see if the model is already loaded
|
|
//
|
|
for (mh=mhHashTable[hash]; mh; mh=mh->next) {
|
|
if (Q_stricmp(mh->name, name) == 0) {
|
|
return mh->handle;
|
|
}
|
|
}
|
|
|
|
// for ( hModel = 1 ; hModel < tr.numModels; hModel++ ) {
|
|
// mod = tr.models[hModel];
|
|
// if ( !strcmp( mod->name, name ) ) {
|
|
// if( mod->type == MOD_BAD ) {
|
|
// return 0;
|
|
// }
|
|
// return hModel;
|
|
// }
|
|
// }
|
|
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
|
|
// allocate a new model_t
|
|
|
|
if ( ( mod = R_AllocModel() ) == NULL ) {
|
|
ri.Printf( PRINT_WARNING, "RE_RegisterModel: R_AllocModel() failed for '%s'\n", name);
|
|
return 0;
|
|
}
|
|
|
|
// only set the name after the model has been successfully loaded
|
|
Q_strncpyz( mod->name, name, sizeof( mod->name ) );
|
|
|
|
#ifndef DEDICATED
|
|
// make sure the render thread is stopped
|
|
R_SyncRenderThread();
|
|
#endif
|
|
|
|
int iLODStart = 0;
|
|
if (strstr (name, ".md3")) {
|
|
iLODStart = MD3_MAX_LODS-1; // this loads the md3s in reverse so they can be biased
|
|
}
|
|
mod->numLods = 0;
|
|
|
|
//
|
|
// load the files
|
|
//
|
|
numLoaded = 0;
|
|
|
|
for ( lod = iLODStart; lod >= 0 ; lod-- ) {
|
|
char filename[1024];
|
|
|
|
strcpy( filename, name );
|
|
|
|
if ( lod != 0 ) {
|
|
char namebuf[80];
|
|
|
|
if ( strrchr( filename, '.' ) ) {
|
|
*strrchr( filename, '.' ) = 0;
|
|
}
|
|
sprintf( namebuf, "_%d.md3", lod );
|
|
strcat( filename, namebuf );
|
|
}
|
|
|
|
qboolean bAlreadyCached = qfalse;
|
|
if (!RE_RegisterModels_GetDiskFile(filename, (void **)&buf, &bAlreadyCached))
|
|
{
|
|
continue;
|
|
}
|
|
|
|
//loadmodel = mod; // this seems to be fairly pointless
|
|
|
|
// important that from now on we pass 'filename' instead of 'name' to all model load functions,
|
|
// because 'filename' accounts for any LOD mangling etc so guarantees unique lookups for yet more
|
|
// internal caching...
|
|
//
|
|
ident = *(unsigned *)buf;
|
|
if (!bAlreadyCached)
|
|
{
|
|
ident = LittleLong(ident);
|
|
}
|
|
|
|
switch (ident)
|
|
{
|
|
// if you add any new types of model load in this switch-case, tell me,
|
|
// or copy what I've done with the cache scheme (-ste).
|
|
//
|
|
case MD4_IDENT:
|
|
|
|
loaded = R_LoadMD4( mod, buf, filename, bAlreadyCached );
|
|
break;
|
|
|
|
case MDXA_IDENT:
|
|
|
|
loaded = R_LoadMDXA( mod, buf, filename, bAlreadyCached );
|
|
break;
|
|
|
|
case MDXM_IDENT:
|
|
|
|
loaded = R_LoadMDXM( mod, buf, filename, bAlreadyCached );
|
|
break;
|
|
|
|
case MD3_IDENT:
|
|
|
|
loaded = R_LoadMD3( mod, lod, buf, filename, bAlreadyCached );
|
|
break;
|
|
|
|
default:
|
|
|
|
ri.Printf (PRINT_WARNING,"RE_RegisterModel: unknown fileid for %s\n", filename);
|
|
goto fail;
|
|
}
|
|
|
|
if (!bAlreadyCached){ // important to check!!
|
|
ri.FS_FreeFile (buf);
|
|
}
|
|
|
|
if ( !loaded ) {
|
|
if ( lod == 0 ) {
|
|
goto fail;
|
|
} else {
|
|
break;
|
|
}
|
|
} else {
|
|
mod->numLods++;
|
|
numLoaded++;
|
|
// if we have a valid model and are biased
|
|
// so that we won't see any higher detail ones,
|
|
// stop loading them
|
|
if ( lod <= r_lodbias->integer ) {
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
if ( numLoaded ) {
|
|
// duplicate into higher lod spots that weren't
|
|
// loaded, in case the user changes r_lodbias on the fly
|
|
for ( lod-- ; lod >= 0 ; lod-- ) {
|
|
mod->numLods++;
|
|
mod->md3[lod] = mod->md3[lod+1];
|
|
}
|
|
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
|
|
RE_InsertModelIntoHash(name, mod);
|
|
return mod->index;
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
}
|
|
#ifdef _DEBUG
|
|
else {
|
|
ri.Printf (PRINT_WARNING,"RE_RegisterModel: couldn't load %s\n", name);
|
|
}
|
|
#endif
|
|
|
|
fail:
|
|
// we still keep the model_t around, so if the model name is asked for
|
|
// again, we won't bother scanning the filesystem
|
|
mod->type = MOD_BAD;
|
|
return 0;
|
|
}
|
|
|
|
|
|
// wrapper function needed to avoid problems with mid-function returns so I can safely use this bool to tell the
|
|
// z_malloc-fail recovery code whether it's safe to ditch any model caches...
|
|
//
|
|
qboolean gbInsideRegisterModel = qfalse;
|
|
qhandle_t RE_RegisterModel( const char *name )
|
|
{
|
|
gbInsideRegisterModel = qtrue; // !!!!!!!!!!!!!!
|
|
|
|
qhandle_t q = RE_RegisterModel_Actual( name );
|
|
|
|
gbInsideRegisterModel = qfalse;
|
|
|
|
return q;
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
=================
|
|
R_LoadMD3
|
|
=================
|
|
*/
|
|
static qboolean R_LoadMD3 (model_t *mod, int lod, void *buffer, const char *mod_name, qboolean bAlreadyCached ) {
|
|
int i, j;
|
|
md3Header_t *pinmodel;
|
|
md3Surface_t *surf;
|
|
md3Shader_t *shader;
|
|
int version;
|
|
int size;
|
|
|
|
#ifndef _M_IX86
|
|
md3Frame_t *frame;
|
|
md3Triangle_t *tri;
|
|
md3St_t *st;
|
|
md3XyzNormal_t *xyz;
|
|
md3Tag_t *tag;
|
|
#endif
|
|
|
|
|
|
pinmodel= (md3Header_t *)buffer;
|
|
//
|
|
// read some fields from the binary, but only LittleLong() them when we know this wasn't an already-cached model...
|
|
//
|
|
version = pinmodel->version;
|
|
size = pinmodel->ofsEnd;
|
|
|
|
if (!bAlreadyCached)
|
|
{
|
|
version = LittleLong(version);
|
|
size = LittleLong(size);
|
|
}
|
|
|
|
if (version != MD3_VERSION) {
|
|
ri.Printf( PRINT_WARNING, "R_LoadMD3: %s has wrong version (%i should be %i)\n",
|
|
mod_name, version, MD3_VERSION);
|
|
return qfalse;
|
|
}
|
|
|
|
mod->type = MOD_MESH;
|
|
mod->dataSize += size;
|
|
|
|
qboolean bAlreadyFound = qfalse;
|
|
mod->md3[lod] = (md3Header_t *) //ri.Hunk_Alloc( size );
|
|
RE_RegisterModels_Malloc(size, mod_name, &bAlreadyFound, TAG_MODEL_MD3);
|
|
|
|
assert(bAlreadyCached == bAlreadyFound); // I should probably eliminate 'bAlreadyFound', but wtf?
|
|
|
|
if (!bAlreadyFound)
|
|
{
|
|
memcpy (mod->md3[lod], buffer, size );
|
|
|
|
LL(mod->md3[lod]->ident);
|
|
LL(mod->md3[lod]->version);
|
|
LL(mod->md3[lod]->numFrames);
|
|
LL(mod->md3[lod]->numTags);
|
|
LL(mod->md3[lod]->numSurfaces);
|
|
LL(mod->md3[lod]->ofsFrames);
|
|
LL(mod->md3[lod]->ofsTags);
|
|
LL(mod->md3[lod]->ofsSurfaces);
|
|
LL(mod->md3[lod]->ofsEnd);
|
|
}
|
|
|
|
if ( mod->md3[lod]->numFrames < 1 ) {
|
|
ri.Printf( PRINT_WARNING, "R_LoadMD3: %s has no frames\n", mod_name );
|
|
return qfalse;
|
|
}
|
|
|
|
if (bAlreadyFound)
|
|
{
|
|
return qtrue; // All done. Stop, go no further, do not pass Go...
|
|
}
|
|
|
|
#ifndef _M_IX86
|
|
//
|
|
// optimisation, we don't bother doing this for standard intel case since our data's already in that format...
|
|
//
|
|
|
|
// swap all the frames
|
|
frame = (md3Frame_t *) ( (byte *)mod->md3[lod] + mod->md3[lod]->ofsFrames );
|
|
for ( i = 0 ; i < mod->md3[lod]->numFrames ; i++, frame++) {
|
|
frame->radius = LittleFloat( frame->radius );
|
|
for ( j = 0 ; j < 3 ; j++ ) {
|
|
frame->bounds[0][j] = LittleFloat( frame->bounds[0][j] );
|
|
frame->bounds[1][j] = LittleFloat( frame->bounds[1][j] );
|
|
frame->localOrigin[j] = LittleFloat( frame->localOrigin[j] );
|
|
}
|
|
}
|
|
|
|
// swap all the tags
|
|
tag = (md3Tag_t *) ( (byte *)mod->md3[lod] + mod->md3[lod]->ofsTags );
|
|
for ( i = 0 ; i < mod->md3[lod]->numTags * mod->md3[lod]->numFrames ; i++, tag++) {
|
|
for ( j = 0 ; j < 3 ; j++ ) {
|
|
tag->origin[j] = LittleFloat( tag->origin[j] );
|
|
tag->axis[0][j] = LittleFloat( tag->axis[0][j] );
|
|
tag->axis[1][j] = LittleFloat( tag->axis[1][j] );
|
|
tag->axis[2][j] = LittleFloat( tag->axis[2][j] );
|
|
}
|
|
}
|
|
#endif
|
|
|
|
// swap all the surfaces
|
|
surf = (md3Surface_t *) ( (byte *)mod->md3[lod] + mod->md3[lod]->ofsSurfaces );
|
|
for ( i = 0 ; i < mod->md3[lod]->numSurfaces ; i++) {
|
|
LL(surf->flags);
|
|
LL(surf->numFrames);
|
|
LL(surf->numShaders);
|
|
LL(surf->numTriangles);
|
|
LL(surf->ofsTriangles);
|
|
LL(surf->numVerts);
|
|
LL(surf->ofsShaders);
|
|
LL(surf->ofsSt);
|
|
LL(surf->ofsXyzNormals);
|
|
LL(surf->ofsEnd);
|
|
|
|
if ( surf->numVerts > SHADER_MAX_VERTEXES ) {
|
|
ri.Error (ERR_DROP, "R_LoadMD3: %s has more than %i verts on a surface (%i)",
|
|
mod_name, SHADER_MAX_VERTEXES, surf->numVerts );
|
|
}
|
|
if ( surf->numTriangles*3 > SHADER_MAX_INDEXES ) {
|
|
ri.Error (ERR_DROP, "R_LoadMD3: %s has more than %i triangles on a surface (%i)",
|
|
mod_name, SHADER_MAX_INDEXES / 3, surf->numTriangles );
|
|
}
|
|
|
|
// change to surface identifier
|
|
surf->ident = SF_MD3;
|
|
|
|
// lowercase the surface name so skin compares are faster
|
|
Q_strlwr( surf->name );
|
|
|
|
// strip off a trailing _1 or _2
|
|
// this is a crutch for q3data being a mess
|
|
j = strlen( surf->name );
|
|
if ( j > 2 && surf->name[j-2] == '_' ) {
|
|
surf->name[j-2] = 0;
|
|
}
|
|
#ifndef DEDICATED
|
|
// register the shaders
|
|
shader = (md3Shader_t *) ( (byte *)surf + surf->ofsShaders );
|
|
for ( j = 0 ; j < surf->numShaders ; j++, shader++ ) {
|
|
shader_t *sh;
|
|
|
|
sh = R_FindShader( shader->name, lightmapsNone, stylesDefault, qtrue );
|
|
if ( sh->defaultShader ) {
|
|
shader->shaderIndex = 0;
|
|
} else {
|
|
shader->shaderIndex = sh->index;
|
|
}
|
|
RE_RegisterModels_StoreShaderRequest(mod_name, &shader->name[0], &shader->shaderIndex);
|
|
}
|
|
#endif
|
|
|
|
#ifndef _M_IX86
|
|
//
|
|
// optimisation, we don't bother doing this for standard intel case since our data's already in that format...
|
|
//
|
|
|
|
// swap all the triangles
|
|
tri = (md3Triangle_t *) ( (byte *)surf + surf->ofsTriangles );
|
|
for ( j = 0 ; j < surf->numTriangles ; j++, tri++ ) {
|
|
LL(tri->indexes[0]);
|
|
LL(tri->indexes[1]);
|
|
LL(tri->indexes[2]);
|
|
}
|
|
|
|
// swap all the ST
|
|
st = (md3St_t *) ( (byte *)surf + surf->ofsSt );
|
|
for ( j = 0 ; j < surf->numVerts ; j++, st++ ) {
|
|
st->st[0] = LittleFloat( st->st[0] );
|
|
st->st[1] = LittleFloat( st->st[1] );
|
|
}
|
|
|
|
// swap all the XyzNormals
|
|
xyz = (md3XyzNormal_t *) ( (byte *)surf + surf->ofsXyzNormals );
|
|
for ( j = 0 ; j < surf->numVerts * surf->numFrames ; j++, xyz++ )
|
|
{
|
|
xyz->xyz[0] = LittleShort( xyz->xyz[0] );
|
|
xyz->xyz[1] = LittleShort( xyz->xyz[1] );
|
|
xyz->xyz[2] = LittleShort( xyz->xyz[2] );
|
|
|
|
xyz->normal = LittleShort( xyz->normal );
|
|
}
|
|
#endif
|
|
|
|
// find the next surface
|
|
surf = (md3Surface_t *)( (byte *)surf + surf->ofsEnd );
|
|
}
|
|
|
|
return qtrue;
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
=================
|
|
R_LoadMD4
|
|
=================
|
|
*/
|
|
static qboolean R_LoadMD4( model_t *mod, void *buffer, const char *mod_name, qboolean bAlreadyCached ) {
|
|
int i;
|
|
md4Header_t *pinmodel, *md4;
|
|
md4LOD_t *lod;
|
|
md4Surface_t *surf;
|
|
|
|
#ifndef _M_IX86
|
|
int j, k;
|
|
md4Frame_t *frame;
|
|
md4Triangle_t *tri;
|
|
md4Vertex_t *v;
|
|
int frameSize;
|
|
#endif
|
|
|
|
int version;
|
|
int size;
|
|
shader_t *sh;
|
|
|
|
pinmodel = (md4Header_t *)buffer;
|
|
//
|
|
// read some fields from the binary, but only LittleLong() them when we know this wasn't an already-cached model...
|
|
//
|
|
version = LittleLong (pinmodel->version);
|
|
size = LittleLong(pinmodel->ofsEnd);
|
|
|
|
if (!bAlreadyCached)
|
|
{
|
|
version = LittleLong(version);
|
|
size = LittleLong(size);
|
|
}
|
|
|
|
if (version != MD4_VERSION) {
|
|
ri.Printf( PRINT_WARNING, "R_LoadMD4: %s has wrong version (%i should be %i)\n",
|
|
mod_name, version, MD4_VERSION);
|
|
return qfalse;
|
|
}
|
|
|
|
mod->type = MOD_MD4;
|
|
mod->dataSize += size;
|
|
|
|
qboolean bAlreadyFound = qfalse;
|
|
md4 = mod->md4 = (md4Header_t *)// ri.Hunk_Alloc( size, h_low );
|
|
RE_RegisterModels_Malloc(size, mod_name, &bAlreadyFound, TAG_MODEL_MD4);
|
|
|
|
assert(bAlreadyCached == bAlreadyFound); // I should probably eliminate 'bAlreadyFound', but wtf?
|
|
|
|
if (!bAlreadyFound)
|
|
{
|
|
Com_Memcpy( md4, buffer, size );
|
|
|
|
LL(md4->ident);
|
|
LL(md4->version);
|
|
LL(md4->numFrames);
|
|
LL(md4->numBones);
|
|
LL(md4->numLODs);
|
|
LL(md4->ofsFrames);
|
|
LL(md4->ofsLODs);
|
|
LL(md4->ofsEnd);
|
|
}
|
|
|
|
mod->numLods = md4->numLODs -1 ; //copy this up to the model for ease of use - it wil get inced after this.
|
|
|
|
if ( md4->numFrames < 1 ) {
|
|
ri.Printf( PRINT_WARNING, "R_LoadMD4: %s has no frames\n", mod_name );
|
|
return qfalse;
|
|
}
|
|
|
|
if (bAlreadyFound)
|
|
{
|
|
return qtrue; // All done. Stop, go no further, do not LittleLong(), do not pass Go...
|
|
}
|
|
|
|
|
|
|
|
// we don't need to swap tags in the renderer, they aren't used
|
|
|
|
#ifndef _M_IX86
|
|
//
|
|
// optimisation, we don't bother doing this for standard intel case since our data's already in that format...
|
|
//
|
|
// swap all the frames
|
|
frameSize = (int)( &((md4Frame_t *)0)->bones[ md4->numBones ] );
|
|
for ( i = 0 ; i < md4->numFrames ; i++, frame++) {
|
|
frame = (md4Frame_t *) ( (byte *)md4 + md4->ofsFrames + i * frameSize );
|
|
frame->radius = LittleFloat( frame->radius );
|
|
for ( j = 0 ; j < 3 ; j++ ) {
|
|
frame->bounds[0][j] = LittleFloat( frame->bounds[0][j] );
|
|
frame->bounds[1][j] = LittleFloat( frame->bounds[1][j] );
|
|
frame->localOrigin[j] = LittleFloat( frame->localOrigin[j] );
|
|
}
|
|
for ( j = 0 ; j < md4->numBones * sizeof( md4Bone_t ) / 4 ; j++ ) {
|
|
((float *)frame->bones)[j] = LittleFloat( ((float *)frame->bones)[j] );
|
|
}
|
|
}
|
|
#endif
|
|
|
|
// swap all the LOD's (we need to do the middle part of this even for intel, because of shader reg and err-check)
|
|
lod = (md4LOD_t *) ( (byte *)md4 + md4->ofsLODs );
|
|
for ( i = 0 ; i < md4->numLODs ; i++) {
|
|
|
|
// swap all the surfaces
|
|
surf = (md4Surface_t *) ( (byte *)lod + lod->ofsSurfaces );
|
|
for ( i = 0 ; i < lod->numSurfaces ; i++) {
|
|
LL(surf->ident);
|
|
LL(surf->numTriangles);
|
|
LL(surf->ofsTriangles);
|
|
LL(surf->numVerts);
|
|
LL(surf->ofsVerts);
|
|
LL(surf->ofsEnd);
|
|
|
|
if ( surf->numVerts > SHADER_MAX_VERTEXES ) {
|
|
ri.Error (ERR_DROP, "R_LoadMD3: %s has more than %i verts on a surface (%i)",
|
|
mod_name, SHADER_MAX_VERTEXES, surf->numVerts );
|
|
}
|
|
if ( surf->numTriangles*3 > SHADER_MAX_INDEXES ) {
|
|
ri.Error (ERR_DROP, "R_LoadMD3: %s has more than %i triangles on a surface (%i)",
|
|
mod_name, SHADER_MAX_INDEXES / 3, surf->numTriangles );
|
|
}
|
|
|
|
// change to surface identifier
|
|
surf->ident = SF_MD4;
|
|
#ifndef DEDICATED
|
|
// register the shaders
|
|
sh = R_FindShader( surf->shader, lightmapsNone, stylesDefault, qtrue );
|
|
if ( sh->defaultShader ) {
|
|
surf->shaderIndex = 0;
|
|
} else {
|
|
surf->shaderIndex = sh->index;
|
|
}
|
|
RE_RegisterModels_StoreShaderRequest(mod_name, &surf->shader[0], &surf->shaderIndex);
|
|
#endif
|
|
|
|
#ifndef _M_IX86
|
|
//
|
|
// optimisation, we don't bother doing this for standard intel case since our data's already in that format...
|
|
//
|
|
|
|
// swap all the triangles
|
|
tri = (md4Triangle_t *) ( (byte *)surf + surf->ofsTriangles );
|
|
for ( j = 0 ; j < surf->numTriangles ; j++, tri++ ) {
|
|
LL(tri->indexes[0]);
|
|
LL(tri->indexes[1]);
|
|
LL(tri->indexes[2]);
|
|
}
|
|
|
|
// swap all the vertexes
|
|
v = (md4Vertex_t *) ( (byte *)surf + surf->ofsVerts );
|
|
for ( j = 0 ; j < surf->numVerts ; j++ ) {
|
|
v->normal[0] = LittleFloat( v->normal[0] );
|
|
v->normal[1] = LittleFloat( v->normal[1] );
|
|
v->normal[2] = LittleFloat( v->normal[2] );
|
|
|
|
v->texCoords[0] = LittleFloat( v->texCoords[0] );
|
|
v->texCoords[1] = LittleFloat( v->texCoords[1] );
|
|
|
|
v->numWeights = LittleLong( v->numWeights );
|
|
|
|
for ( k = 0 ; k < v->numWeights ; k++ ) {
|
|
v->weights[k].boneIndex = LittleLong( v->weights[k].boneIndex );
|
|
v->weights[k].boneWeight = LittleFloat( v->weights[k].boneWeight );
|
|
v->weights[k].offset[0] = LittleFloat( v->weights[k].offset[0] );
|
|
v->weights[k].offset[1] = LittleFloat( v->weights[k].offset[1] );
|
|
v->weights[k].offset[2] = LittleFloat( v->weights[k].offset[2] );
|
|
}
|
|
v = (md4Vertex_t *)&v->weights[v->numWeights];
|
|
}
|
|
#endif
|
|
|
|
// find the next surface
|
|
surf = (md4Surface_t *)( (byte *)surf + surf->ofsEnd );
|
|
}
|
|
|
|
// find the next LOD
|
|
lod = (md4LOD_t *)( (byte *)lod + lod->ofsEnd );
|
|
}
|
|
|
|
return qtrue;
|
|
}
|
|
|
|
|
|
//=============================================================================
|
|
#ifndef DEDICATED
|
|
/*
|
|
** RE_BeginRegistration
|
|
*/
|
|
void RE_BeginRegistration( glconfig_t *glconfigOut ) {
|
|
|
|
R_Init();
|
|
|
|
*glconfigOut = glConfig;
|
|
|
|
R_SyncRenderThread();
|
|
|
|
tr.viewCluster = -1; // force markleafs to regenerate
|
|
|
|
// rww - 9-13-01 [1-26-01-sof2]
|
|
//R_ClearFlares();
|
|
|
|
RE_ClearScene();
|
|
|
|
tr.registered = qtrue;
|
|
|
|
// NOTE: this sucks, for some reason the first stretch pic is never drawn
|
|
// without this we'd see a white flash on a level load because the very
|
|
// first time the level shot would not be drawn
|
|
RE_StretchPic(0, 0, 0, 0, 0, 0, 1, 1, 0);
|
|
}
|
|
|
|
//=============================================================================
|
|
|
|
#endif // !DEDICATED
|
|
void R_SVModelInit()
|
|
{
|
|
model_t *mod;
|
|
|
|
tr.numModels = 0;
|
|
memset(mhHashTable, 0, sizeof(mhHashTable));
|
|
|
|
mod = GenericAllocModel();
|
|
mod->type = MOD_BAD;
|
|
}
|
|
|
|
/*
|
|
===============
|
|
R_ModelInit
|
|
===============
|
|
*/
|
|
void R_ModelInit( void ) {
|
|
model_t *mod;
|
|
|
|
// leave a space for NULL model
|
|
tr.numModels = 0;
|
|
/*
|
|
Ghoul2 Insert Start
|
|
*/
|
|
|
|
memset(mhHashTable, 0, sizeof(mhHashTable));
|
|
/*
|
|
Ghoul2 Insert End
|
|
*/
|
|
|
|
mod = R_AllocModel();
|
|
mod->type = MOD_BAD;
|
|
}
|
|
|
|
|
|
/*
|
|
================
|
|
R_Modellist_f
|
|
================
|
|
*/
|
|
void R_Modellist_f( void ) {
|
|
int i, j;
|
|
model_t *mod;
|
|
int total;
|
|
int lods;
|
|
|
|
total = 0;
|
|
for ( i = 1 ; i < tr.numModels; i++ ) {
|
|
mod = tr.models[i];
|
|
lods = 1;
|
|
for ( j = 1 ; j < MD3_MAX_LODS ; j++ ) {
|
|
if ( mod->md3[j] && mod->md3[j] != mod->md3[j-1] ) {
|
|
lods++;
|
|
}
|
|
}
|
|
ri.Printf( PRINT_ALL, "%8i : (%i) %s\n",mod->dataSize, lods, mod->name );
|
|
total += mod->dataSize;
|
|
}
|
|
ri.Printf( PRINT_ALL, "%8i : Total models\n", total );
|
|
|
|
#if 0 // not working right with new hunk
|
|
if ( tr.world ) {
|
|
ri.Printf( PRINT_ALL, "\n%8i : %s\n", tr.world->dataSize, tr.world->name );
|
|
}
|
|
#endif
|
|
}
|
|
|
|
|
|
//=============================================================================
|
|
|
|
|
|
/*
|
|
================
|
|
R_GetTag
|
|
================
|
|
*/
|
|
static md3Tag_t *R_GetTag( md3Header_t *mod, int frame, const char *tagName ) {
|
|
md3Tag_t *tag;
|
|
int i;
|
|
|
|
if ( frame >= mod->numFrames ) {
|
|
// it is possible to have a bad frame while changing models, so don't error
|
|
frame = mod->numFrames - 1;
|
|
}
|
|
|
|
tag = (md3Tag_t *)((byte *)mod + mod->ofsTags) + frame * mod->numTags;
|
|
for ( i = 0 ; i < mod->numTags ; i++, tag++ ) {
|
|
if ( !strcmp( tag->name, tagName ) ) {
|
|
return tag; // found it
|
|
}
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
|
|
/*
|
|
================
|
|
R_LerpTag
|
|
================
|
|
*/
|
|
int R_LerpTag( orientation_t *tag, qhandle_t handle, int startFrame, int endFrame,
|
|
float frac, const char *tagName ) {
|
|
md3Tag_t *start, *end;
|
|
int i;
|
|
float frontLerp, backLerp;
|
|
model_t *model;
|
|
|
|
model = R_GetModelByHandle( handle );
|
|
if ( !model->md3[0] ) {
|
|
AxisClear( tag->axis );
|
|
VectorClear( tag->origin );
|
|
return qfalse;
|
|
}
|
|
|
|
start = R_GetTag( model->md3[0], startFrame, tagName );
|
|
end = R_GetTag( model->md3[0], endFrame, tagName );
|
|
if ( !start || !end ) {
|
|
AxisClear( tag->axis );
|
|
VectorClear( tag->origin );
|
|
return qfalse;
|
|
}
|
|
|
|
frontLerp = frac;
|
|
backLerp = 1.0f - frac;
|
|
|
|
for ( i = 0 ; i < 3 ; i++ ) {
|
|
tag->origin[i] = start->origin[i] * backLerp + end->origin[i] * frontLerp;
|
|
tag->axis[0][i] = start->axis[0][i] * backLerp + end->axis[0][i] * frontLerp;
|
|
tag->axis[1][i] = start->axis[1][i] * backLerp + end->axis[1][i] * frontLerp;
|
|
tag->axis[2][i] = start->axis[2][i] * backLerp + end->axis[2][i] * frontLerp;
|
|
}
|
|
VectorNormalize( tag->axis[0] );
|
|
VectorNormalize( tag->axis[1] );
|
|
VectorNormalize( tag->axis[2] );
|
|
return qtrue;
|
|
}
|
|
|
|
|
|
/*
|
|
====================
|
|
R_ModelBounds
|
|
====================
|
|
*/
|
|
void R_ModelBounds( qhandle_t handle, vec3_t mins, vec3_t maxs ) {
|
|
model_t *model;
|
|
md3Header_t *header;
|
|
md3Frame_t *frame;
|
|
|
|
model = R_GetModelByHandle( handle );
|
|
|
|
if ( model->bmodel ) {
|
|
VectorCopy( model->bmodel->bounds[0], mins );
|
|
VectorCopy( model->bmodel->bounds[1], maxs );
|
|
return;
|
|
}
|
|
|
|
if ( !model->md3[0] ) {
|
|
VectorClear( mins );
|
|
VectorClear( maxs );
|
|
return;
|
|
}
|
|
|
|
header = model->md3[0];
|
|
|
|
frame = (md3Frame_t *)( (byte *)header + header->ofsFrames );
|
|
|
|
VectorCopy( frame->bounds[0], mins );
|
|
VectorCopy( frame->bounds[1], maxs );
|
|
}
|
|
|
|
|