November 2020 - NUT-devel - lists.mplayerhq.hu

[PATCH] Add st_sd_displaymatrix for storing the stream display matrix
by Matthias Neugebauer 13 Dec '20

13 Dec '20

See FFmpeg patch: https://patchwork.ffmpeg.org/project/ffmpeg/patch/F7848CEE-46C1-4CAF-9191-A… --- docs/nut.txt | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/docs/nut.txt b/docs/nut.txt index 3641c01..ea49eba 100644 --- a/docs/nut.txt +++ b/docs/nut.txt @@ -1,5 +1,5 @@ ================================== -NUT Open Container Format 20201104 +NUT Open Container Format 20201120 ================================== @@ -1045,6 +1045,12 @@ info packet types framerates in the stream). Note, this value is just a guess! For example, if the time base is 1/90000 and all frames have either approximately 3600 or 1800 timer ticks, then r_frame_rate will be 50/1. + "st_sd_displaymatrix" + Display matrix of the stream. + This stores the 3x3 display matrix of the stream as nine unsigned + integers seperated by colons. The display matrix is a side data of a + video stream and necessary for displaying the stream correctly, e.g. + rotating it. Note: Nonstandard fields should be prefixed by "X-". value -- 2.29.2

2 1

[nut]: r695 - docs/nut.txt
by michael 06 Nov '20

06 Nov '20

Author: michael Date: Fri Nov 6 23:22:24 2020 New Revision: 695 Log: Add r_frame_rate, which has been stored without X- prefix for a long time The description is taken from the API documentation of libavformat Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt Sun Feb 2 12:47:56 2020 (r694) +++ docs/nut.txt Fri Nov 6 23:22:24 2020 (r695) @@ -1,5 +1,5 @@ ================================== -NUT Open Container Format 20200130 +NUT Open Container Format 20201104 ================================== @@ -1038,6 +1038,13 @@ info packet types instead? is there some other unique way to identify fonts? This stream uses the specified font, if it is unavailable presentation might be less pretty. + "r_frame_rate" + Real base framerate of the stream. + This is the lowest framerate with which all timestamps can be + represented accurately (it is the least common multiple of all + framerates in the stream). Note, this value is just a guess! + For example, if the time base is 1/90000 and all frames have either + approximately 3600 or 1800 timer ticks, then r_frame_rate will be 50/1. Note: Nonstandard fields should be prefixed by "X-". value

1 0

[PATCH 1/7] [RFC] Revert "avcodec/adpcm_swf: support decoding multiple fixed-sized blocks at once"
by Michael Niedermayer 05 Nov '20

05 Nov '20

The reverted code split at block_align boundaries, but there was already code which splits at a hardcoded 4096 sample boundary. reverting this seemed like the easiest fix but this is a RFC in case another solution is preferred Fixes: out of array write Fixes: 26821/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ADPCM_SWF_fuzzer-5764465137811456 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg This reverts commit e9dd73d30d09043446ac6dd7b8ad31e557873852. --- libavcodec/adpcm.c | 15 ++------------- 1 file changed, 2 insertions(+), 13 deletions(-) diff --git a/libavcodec/adpcm.c b/libavcodec/adpcm.c index 701b125c47..d018c1f91b 100644 --- a/libavcodec/adpcm.c +++ b/libavcodec/adpcm.c @@ -880,7 +880,7 @@ static int get_nb_samples(AVCodecContext *avctx, GetByteContext *gb, } case AV_CODEC_ID_ADPCM_SWF: { - int buf_bits = (avctx->block_align ? avctx->block_align : buf_size) * 8 - 2; + int buf_bits = buf_size * 8 - 2; int nbits = (bytestream2_get_byte(gb) >> 6) + 2; int block_hdr_size = 22 * ch; int block_size = block_hdr_size + nbits * ch * 4095; @@ -889,9 +889,6 @@ static int get_nb_samples(AVCodecContext *avctx, GetByteContext *gb, nb_samples = nblocks * 4096; if (bits_left >= block_hdr_size) nb_samples += 1 + (bits_left - block_hdr_size) / (nbits * ch); - - if (avctx->block_align) - nb_samples *= buf_size / avctx->block_align; break; } case AV_CODEC_ID_ADPCM_THP: @@ -1770,17 +1767,9 @@ static int adpcm_decode_frame(AVCodecContext *avctx, void *data, } break; case AV_CODEC_ID_ADPCM_SWF: - { - const int nb_blocks = avctx->block_align ? avpkt->size / avctx->block_align : 1; - const int block_size = avctx->block_align ? avctx->block_align : avpkt->size; - - for (int block = 0; block < nb_blocks; block++) { - adpcm_swf_decode(avctx, buf + block * block_size, block_size, samples); - samples += nb_samples / nb_blocks; - } + adpcm_swf_decode(avctx, buf, buf_size, samples); bytestream2_seek(&gb, 0, SEEK_END); break; - } case AV_CODEC_ID_ADPCM_YAMAHA: for (n = nb_samples >> (1 - st); n > 0; n--) { int v = bytestream2_get_byteu(&gb); -- 2.17.1

1 6

[PATCH] Add r_frame_rate, which has been stored without X- prefix for a long time
by Michael Niedermayer 04 Nov '20

04 Nov '20

The description is taken from the API documentation of libavformat --- docs/nut.txt | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/docs/nut.txt b/docs/nut.txt index 2cccb51..3641c01 100644 --- a/docs/nut.txt +++ b/docs/nut.txt @@ -1,5 +1,5 @@ ================================== -NUT Open Container Format 20200130 +NUT Open Container Format 20201104 ================================== @@ -1038,6 +1038,13 @@ info packet types instead? is there some other unique way to identify fonts? This stream uses the specified font, if it is unavailable presentation might be less pretty. + "r_frame_rate" + Real base framerate of the stream. + This is the lowest framerate with which all timestamps can be + represented accurately (it is the least common multiple of all + framerates in the stream). Note, this value is just a guess! + For example, if the time base is 1/90000 and all frames have either + approximately 3600 or 1800 timer ticks, then r_frame_rate will be 50/1. Note: Nonstandard fields should be prefixed by "X-". value -- 2.17.1

1 0