Dear Maintainer, Debian is maintaining this right? Upsteam seams dead. I recently compiled tuxkart (uses libplib), not supertuxkart, tuxkart. It is the ancestor of supertuxkart. I was amazed to find out that it actually worked O_O (2006) . The only problem, is that it compiled with OSS support. And it's a version that is not compatible with the simple OSS emulations that just load a library, but require full OSS emulation with a fake /dev/dsp. Trying to emulate that is especially inefficient and buggy. the problem is in the SL library inside libplib. It would require to be updated with pulseaudio. Asuming this is not a wontfix.
Upstream is sadly dead. Steve Baker (upstream dev) has abandoned PLib completely <https://sourceforge.net/p/plib/bugs/55/#88e9>. I am working on a project to bring back TuxKart on newer Debian machines, and I also came across this issue. It seems that the PLib git repo (on Salsa) is not really being updated much, besides for some patches. alsa-oss and other OSS emulators don't work, so it would be nice to have an ALSA patch on PLib. On Sat, 12 Aug 2023 01:32:07 +0200 michel <okgomdjgbmoij@gmail.com> wrote: > Package: libplib1 > Version: 1.8.5-14+b1 > Severity: normal > X-Debbugs-Cc: okgomdjgbmoij@gmail.com > > Dear Maintainer, > > > > Debian is maintaining this right? Upsteam seams dead. > > I recently compiled tuxkart (uses libplib), not supertuxkart, tuxkart. It is the ancestor of supertuxkart. > > I was amazed to find out that it actually worked O_O (2006) . The only problem, is that it compiled with OSS support. And it's a version that is not compatible with the simple OSS emulations that just load a library, but require full OSS emulation with a fake /dev/dsp. Trying to emulate that is especially inefficient and buggy. > > the problem is in the SL library inside libplib. It would require to be updated with pulseaudio. > Asuming this is not a wontfix. > >
Dear Maintainer Here's a patch with pulseaudio support (applied last). I made it with AI and don't understand exactly what's it's doing, so don't bother to ask me. I did my best to avoid the AI weirdness. I did design decisions, the AI wasn't able to do it on it's own. I'm amazed that it works. On a high level, it has a thin wrapper as the expected sound interface of the library and a pulseaudio manager class with a thread as the backend. I split it this way so that there's only one pulseaudio managing object and avoid synchronization and blocking writes issues. it works on my slow laptop, maybe on a faster machine other synchronization bugs will show up. the target application for the fix is tuxkart, the ancestor of supertuxkart. You can easily build it with makedeb on the mpr (it's like the AUR but for debs) https://mpr.makedeb.org/packages/tuxkart i'm not sure how exactly to package the lib the debian way, i'm using this to run configure. I didn't touch the building process, it requires the pulseaudio flags to use the patch CXXFLAGS="-fPIC -DSL_USE_PULSEAUDIO" CFLAGS="-fPIC -DSL_USE_PULSEAUDIO" LDFLAGS="-fPIC -lpulse-simple -lpulse -lpthread" ./configure --prefix=/usr