Patches attached. - Andreas
From b58c71c2e90380d8a3e7a58fa20a6c16c2458317 Mon Sep 17 00:00:00 2001 From: Andreas Rheinhardt <andreas.rheinha...@outlook.com> Date: Sun, 25 May 2025 03:31:17 +0200 Subject: [PATCH 1/3] avformat/matroska: Support JPEG2000 for demuxing
Legal since commit 1cd0a9be4b2d1e7c60184ec68404e00e46e3123e (Jan 4) in the Cellar Matroska specification git repo. We still hold out on muxing it due to compatibility with old demuxers. Signed-off-by: Andreas Rheinhardt <andreas.rheinha...@outlook.com> --- libavformat/matroska.c | 1 + libavformat/matroskaenc.c | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/libavformat/matroska.c b/libavformat/matroska.c index bbad9a7f54..60584e2687 100644 --- a/libavformat/matroska.c +++ b/libavformat/matroska.c @@ -82,6 +82,7 @@ const CodecTags ff_mkv_codec_tags[]={ {"V_AVS3" , AV_CODEC_ID_AVS3}, {"V_DIRAC" , AV_CODEC_ID_DIRAC}, {"V_FFV1" , AV_CODEC_ID_FFV1}, + {"V_JPEG2000" , AV_CODEC_ID_JPEG2000}, {"V_MJPEG" , AV_CODEC_ID_MJPEG}, {"V_MPEG1" , AV_CODEC_ID_MPEG1VIDEO}, {"V_MPEG2" , AV_CODEC_ID_MPEG2VIDEO}, diff --git a/libavformat/matroskaenc.c b/libavformat/matroskaenc.c index 6d0d791f18..9d13f74907 100644 --- a/libavformat/matroskaenc.c +++ b/libavformat/matroskaenc.c @@ -1960,8 +1960,8 @@ static int mkv_write_track(AVFormatContext *s, MatroskaMuxContext *mkv, // look for a codec ID string specific to mkv to use, // if none are found, use AVI codes - if (par->codec_id == AV_CODEC_ID_FFV1) { - /* FFV1 is actually supported natively in Matroska, + if (par->codec_id == AV_CODEC_ID_FFV1 || par->codec_id == AV_CODEC_ID_JPEG2000) { + /* FFV1 and JPEG2000 are actually supported natively in Matroska, * yet we use the VfW way to mux it for compatibility * with old demuxers. (FIXME: Are they really important?) */ } else if (par->codec_id != AV_CODEC_ID_RAWVIDEO || par->codec_tag) { -- 2.45.2
From 5d8954294fee0d1086047ae2587856c3afb7ff7c Mon Sep 17 00:00:00 2001 From: Andreas Rheinhardt <andreas.rheinha...@outlook.com> Date: Sun, 25 May 2025 05:12:05 +0200 Subject: [PATCH 2/3] avformat/matroskadec: Fix VfW extradata size The structure is padded to an even length with an internal size field to indicate the real size. The matroska-matroska-display-metadata test (writing FFV1 in VFW mode) was affected by this. Signed-off-by: Andreas Rheinhardt <andreas.rheinha...@outlook.com> --- libavformat/matroskadec.c | 5 +++++ tests/ref/fate/matroska-mastering-display-metadata | 2 +- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c index 2750652c51..da5166319e 100644 --- a/libavformat/matroskadec.c +++ b/libavformat/matroskadec.c @@ -2877,6 +2877,11 @@ static int mkv_parse_video_codec(MatroskaTrack *track, AVCodecParameters *par, { if (!strcmp(track->codec_id, "V_MS/VFW/FOURCC") && track->codec_priv.size >= 40) { + uint32_t size = AV_RL32A(track->codec_priv.data); + // VFW extradata is padded to an even length, yet + // the size field contains the real length. + if (size & 1 && size == track->codec_priv.size - 1) + --track->codec_priv.size; track->ms_compat = 1; par->bits_per_coded_sample = AV_RL16(track->codec_priv.data + 14); par->codec_tag = AV_RL32(track->codec_priv.data + 16); diff --git a/tests/ref/fate/matroska-mastering-display-metadata b/tests/ref/fate/matroska-mastering-display-metadata index 6a2ff15b1b..5984f54131 100644 --- a/tests/ref/fate/matroska-mastering-display-metadata +++ b/tests/ref/fate/matroska-mastering-display-metadata @@ -1,7 +1,7 @@ c1e5e2ecf433cf05af8556debc7d4d0b *tests/data/fate/matroska-mastering-display-metadata.matroska 1669773 tests/data/fate/matroska-mastering-display-metadata.matroska #extradata 0: 4, 0x040901a3 -#extradata 3: 202, 0xfce96279 +#extradata 3: 201, 0x9a706279 #tb 0: 1/1000 #media_type 0: video #codec_id 0: prores -- 2.45.2
From 5ac5da0a6e97304b678c08d7047d3b19f12817ac Mon Sep 17 00:00:00 2001 From: Andreas Rheinhardt <andreas.rheinha...@outlook.com> Date: Sun, 25 May 2025 04:26:20 +0200 Subject: [PATCH 3/3] avformat/matroskaenc: Use native id instead of VfW for FFV1 Up until now, our muxer wrote FFV1 in video-for-windows compatibility mode out of concern for old demuxers that only support that (whereas the demuxer accepts V_FFV1). This commit switches to using native mode, because a) V_FFV1 is around long enough so that old demuxers should not be an issue (support in FFmpeg has been added in commit 9ae762da7e256aa4d3b645c614fcd1959e1cbb8d in March 2017/FFmpeg 3.3), b) using native mode uses fewer bytes for the CodecPrivate, c) the VfW extradata is zero-padded to an even length if necessary, but our demuxer forgot to undo the padding until very recently, so that there are many versions of our demuxer around that are buggy wrt VFW, but not V_FFV1. This affects the FFV1 extradata checksums, specifically the (experimental) version 4 files with error check version 2* as created by ffmpeg -i ../fate-suite/mpeg2/sony-ct3.bs -c:v ffv1 \ -slices 16 -frames 1 -level 4 -strict experimental ffv1.mkv VFW files like the above created by this muxer before this patch would not work with an old demuxer. *: without error check version 2, the CRC for the whole extradata is zero, which is not changed by appending a zero byte. Signed-off-by: Andreas Rheinhardt <andreas.rheinha...@outlook.com> --- libavformat/matroskaenc.c | 4 ++-- tests/ref/fate/matroska-mastering-display-metadata | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/libavformat/matroskaenc.c b/libavformat/matroskaenc.c index 9d13f74907..408890fa89 100644 --- a/libavformat/matroskaenc.c +++ b/libavformat/matroskaenc.c @@ -1960,8 +1960,8 @@ static int mkv_write_track(AVFormatContext *s, MatroskaMuxContext *mkv, // look for a codec ID string specific to mkv to use, // if none are found, use AVI codes - if (par->codec_id == AV_CODEC_ID_FFV1 || par->codec_id == AV_CODEC_ID_JPEG2000) { - /* FFV1 and JPEG2000 are actually supported natively in Matroska, + if (par->codec_id == AV_CODEC_ID_JPEG2000) { + /* JPEG2000 is actually supported natively in Matroska, * yet we use the VfW way to mux it for compatibility * with old demuxers. (FIXME: Are they really important?) */ } else if (par->codec_id != AV_CODEC_ID_RAWVIDEO || par->codec_tag) { diff --git a/tests/ref/fate/matroska-mastering-display-metadata b/tests/ref/fate/matroska-mastering-display-metadata index 5984f54131..6f10dc57a6 100644 --- a/tests/ref/fate/matroska-mastering-display-metadata +++ b/tests/ref/fate/matroska-mastering-display-metadata @@ -1,5 +1,5 @@ -c1e5e2ecf433cf05af8556debc7d4d0b *tests/data/fate/matroska-mastering-display-metadata.matroska -1669773 tests/data/fate/matroska-mastering-display-metadata.matroska +bdca53906b34c57192416a0f737b885e *tests/data/fate/matroska-mastering-display-metadata.matroska +1669723 tests/data/fate/matroska-mastering-display-metadata.matroska #extradata 0: 4, 0x040901a3 #extradata 3: 201, 0x9a706279 #tb 0: 1/1000 -- 2.45.2
_______________________________________________ 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".