* The hash of textures is now computed the first time they are used. The hash is recomputed if the content might have changed.
* In some games, textures are used as rendering targets (e.g. a mirror or a portal in Portal). The hash calculation of these textures might slow down the game.
-> there is an option to disable the computation of hashes for rendering targets.
* Enable or Disable the support of tpf mods affect now the game immediately.
* Little rearrangements in the GUI.
* uMod can now extract the content of mods to disk and the game loads the textures from there. This solves the huge memory problem of uMod.
This commit is contained in:
code@koerner-de.net
2013-01-07 07:43:38 +00:00
parent 2593729634
commit 601c8c1504
34 changed files with 1523 additions and 1237 deletions
+12 -11
View File
@@ -28,7 +28,6 @@ uMod_TextureClient::uMod_TextureClient(const int version) : Version(version)
BoolSaveAllTextures = false;
BoolSaveSingleTexture = false;
BoolShowTextureString = false;
BoolComputeCRC = false;
KeyBack = 0;
KeySave = 0;
@@ -63,11 +62,13 @@ uMod_TextureClient::~uMod_TextureClient(void)
if (Mutex!=NULL) CloseHandle(Mutex);
if (Update!=NULL) delete [] Update;
if (FileToMod!=NULL)
if (FileToMod!=NULL) delete [] FileToMod;
/*
{
for (int i=0; i<NumberToMod; i++) if (FileToMod[i].Textures!=NULL) delete [] FileToMod[i].Textures;
delete [] FileToMod;
}
*/
}
@@ -121,7 +122,7 @@ int uMod_TextureClient::SetGameName( wchar_t *name)
int uMod_TextureClient::AddUpdate(TextureFileStruct* update, int number) //client must delete the update array
int uMod_TextureClient::AddUpdate(TextureEntry* update, int number) //client must delete the update array
{
Message("AddUpdate( %p, %d): %p\n", update, number, this);
if (int ret = LockMutex()) {gl_ErrorState |= uMod_ERROR_TEXTURE; return (ret);}
@@ -152,7 +153,7 @@ int uMod_TextureClient::GetIndex( DWORD64 hash, int num_index_list, int *index_l
{
if (index_list==NULL || num_index_list==0)
{
if (hash<FileToMod[0].Hash || hash>FileToMod[NumberToMod-1].Hash) return (-1);
if (hash<FileToMod[0].Content()->Hash || hash>FileToMod[NumberToMod-1].Content()->Hash) return (-1);
int pos = NumberToMod/2;
int begin = 0;
int end = NumberToMod-1;
@@ -162,23 +163,23 @@ int uMod_TextureClient::GetIndex( DWORD64 hash, int num_index_list, int *index_l
// Note: contradicting to normal C-code here the interval includes the index "begin" and "end"!
while (begin+1<end) // as long as the interval is longer than two
{
if (hash > FileToMod[pos].Hash) // the new interval is the right half of the actual interval
if (hash > FileToMod[pos].Content()->Hash) // the new interval is the right half of the actual interval
{
begin = pos+1; // the new interval does not contain the index "pos"
pos = (begin + end)/2; // set "pos" somewhere inside the new interval
}
else if (hash < FileToMod[pos].Hash) // the new interval is the left half of the actual interval
else if (hash < FileToMod[pos].Content()->Hash) // the new interval is the left half of the actual interval
{
end = pos-1; // the new interval does not contain the index "pos"
pos = (begin + end)/2; // set "pos" somewhere inside the new interval
}
else {return (pos); break;} // we hit the correct hash
}
for ( pos=begin; pos<=end; pos++) if (FileToMod[pos].Hash==hash) return (pos);
for ( pos=begin; pos<=end; pos++) if (FileToMod[pos].Content()->Hash==hash) return (pos);
}
else
{
if (hash<FileToMod[index_list[0]].Hash || hash>FileToMod[index_list[num_index_list-1]].Hash) return (-1);
if (hash<FileToMod[index_list[0]].Content()->Hash || hash>FileToMod[index_list[num_index_list-1]].Content()->Hash) return (-1);
int pos = num_index_list/2;
int begin = 0;
int end = num_index_list-1;
@@ -188,19 +189,19 @@ int uMod_TextureClient::GetIndex( DWORD64 hash, int num_index_list, int *index_l
// Note: contradicting to normal C-code here the interval includes the index "begin" and "end"!
while (begin+1<end) // as long as the interval is longer than two
{
if (hash > FileToMod[index_list[pos]].Hash) // the new interval is the right half of the actual interval
if (hash > FileToMod[index_list[pos]].Content()->Hash) // the new interval is the right half of the actual interval
{
begin = pos+1; // the new interval does not contain the index "pos"
pos = (begin + end)/2; // set "pos" somewhere inside the new interval
}
else if (hash < FileToMod[index_list[pos]].Hash) // the new interval is the left half of the actual interval
else if (hash < FileToMod[index_list[pos]].Content()->Hash) // the new interval is the left half of the actual interval
{
end = pos-1; // the new interval does not contain the index "pos"
pos = (begin + end)/2; // set "pos" somewhere inside the new interval
}
else {return (index_list[pos]); break;} // we hit the correct hash
}
for ( pos=begin; pos<=end; pos++) if (FileToMod[index_list[pos]].Hash==hash) return (index_list[pos]);
for ( pos=begin; pos<=end; pos++) if (FileToMod[index_list[pos]].Content()->Hash==hash) return (index_list[pos]);
}
}
return (-1);