you probably want to make a pr over at https://code.ffmpeg.org
On Fri, Feb 20, 2026 at 8:41 PM marcos ashton via ffmpeg-devel < [email protected]> wrote: > Hi, this is my qualification task for the GSoC 2026 > "Improving FATE Test Coverage" project. > > This series adds 7 new FATE tests for three previously untested files: > > 1/3: Unit test for libavutil/film_grain_params.c (~98% coverage) > 2/3: Encode/decode roundtrip for libavcodec/cngdec.c (~92% coverage) > 3/3: Filter tests for libavfilter/af_aderivative.c (~87% coverage) > > marcos ashton (3): > tests/fate/libavutil: add FATE test for film_grain_params > tests/fate/voice: add FATE test for CNG encoder roundtrip > tests/fate/filter-audio: add FATE tests for aderivative and aintegral > > libavutil/Makefile | 1 + > libavutil/tests/film_grain_params.c | 240 +++++++++++++++++++++++++ > tests/fate/filter-audio.mak | 25 +++ > tests/fate/libavutil.mak | 4 + > tests/fate/voice.mak | 8 + > tests/ref/fate/cng-encode | 84 +++++++++ > tests/ref/fate/film_grain_params | 30 ++++ > tests/ref/fate/filter-aderivative | 70 ++++++++ > tests/ref/fate/filter-aderivative-fltp | 70 ++++++++ > tests/ref/fate/filter-aderivative-s32p | 70 ++++++++ > tests/ref/fate/filter-aintegral | 70 ++++++++ > tests/ref/fate/filter-aintegral-dblp | 70 ++++++++ > 12 files changed, 742 insertions(+) > create mode 100644 libavutil/tests/film_grain_params.c > create mode 100644 tests/ref/fate/cng-encode > create mode 100644 tests/ref/fate/film_grain_params > create mode 100644 tests/ref/fate/filter-aderivative > create mode 100644 tests/ref/fate/filter-aderivative-fltp > create mode 100644 tests/ref/fate/filter-aderivative-s32p > create mode 100644 tests/ref/fate/filter-aintegral > create mode 100644 tests/ref/fate/filter-aintegral-dblp > > -- > 2.53.0 > > _______________________________________________ > ffmpeg-devel mailing list -- [email protected] > To unsubscribe send an email to [email protected] > _______________________________________________ ffmpeg-devel mailing list -- [email protected] To unsubscribe send an email to [email protected]
