one pointer arithmetic one in TextureFilters.cpp (see tracker).
Once that is fixed Issue 24 in current form can be closed.
Testing done: Code compiles for normal and debug use without
any -Wall warnings and seems to yield the same results and previous
svn on my r300.
Rice Video as part of compiler warning code cleanup (Issue 24).
Testing done: Code compiles for normal and debug use without
general level warning and seems to yield the same results as previous
svn on my r300.
-Added new tab to configuration dialog for user to create mappings.
-Minor fix: Reset configuration dialog to have first tab selected everytime it's shown.
-Added file filters to load dialogs for a more polished file chooser.
-Removed chdir to the config dir (~/.mupen64plus) from setPaths function. This was a hack to get around the fact that plugins were looking in the current working dir for config files. I've since fixed that issue, so the chdir is no longer necessary. Plus, the chdir was causing the file chooser dialogs to default to ~/.mupen64plus, which could be confusing to a user.
JTTL's Audio Plugin
- Fixed name output, in some places it was "Sound" before.
- Improved debug output and used Rice Video conventions for printf (the "[plugin name] message" convention)
- Added Libsamplerate support
- Bug fix: Loads configuration before setting frequency (Before the frequency in the config file was not used)
- Added new options, RESAMPLE and TIME_COMPENSATION. Both have options 0-2, see updated jttl_audio.conf for more information.
- Added new resampler (libsamplerate Best Quality SINC resampler), established options for the others (None, and the default, Linear resampling)
- Better GTK Test output.
- Fixed jttl_audio/main.c indentation, because it was indented differently throughout the file. (That will probably make this patch way bigger than it should be. Sorry.)
I will try to do more work on the GUI of JTTL, and maybe add more resampler options. And if anyone can figure out the problem with libresamplerate's resampler causing problems (I'm aware its a memory overflow problem, But I can't quite figure out what to do to fix it, since somehow it manages not to crash anything running, but if I do any of the things that did fix the segfaulting on exit the buffer did not get filled all the way during gameplay. Its an ugly situation and its definitally my fault.) The new resampler is not default, for its problems and its unoptimized nature (I'm not using the streaming APIs of libsamplerate, so it can probably be further optimized.)
Well, Here goes. I'll cross my fingers and hope that nothing was broken in the process of this patch.
