On Tue, May 06, 2025 at 01:06:40AM +0200, Michael Niedermayer wrote: > This patch set will remove libpostproc from FFmpeg > resulting in a master branch like: > https://github.com/michaelni/FFmpeg/tree/experiment-sourceplugin-master > > matching this, libpostproc can trivially become a source plugin like here: > https://github.com/michaelni/FFmpeg/tree/experiment-sourceplugin-libpostproc > > git merge between the 2 will simply put libpostproc back. Both sides > can evolve and merge will still work fine. Conflicts would only be > expected if changes cross each other. > > This is in fact so simple its almost hard to believe > > Also, I have split out libpostproc in a more conventional style > here: https://github.com/michaelni/libpostproc > This also has been updated already once to include changes from > 5months of changes to the build system, libavutil and fate. > > Both aprouches can be compared but so far the source plugin style > is less work and i expect it to continue to be easier. > > The idea of course here is to expand this to filters and other > things. Which again is trivial, nothing really is needed except > people simply following this style of a source plugin > > What will make source plugins even easier is if the files > that they touch, like allfilters/allcodecs/Makefile/... would be split > so conflicts change from rare to impossible. > > Note, in case this is applied before 8.0, I do intend to include > the libpostproc source plugin in the 8.0 release, so as not to > cause more delays or inconveniences
patchset applied [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Does the universe only have a finite lifespan? No, its going to go on forever, its just that you wont like living in it. -- Hiranya Peiri
signature.asc
Description: PGP signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
