On 01/04/2016 10:34 PM, Hendrik Leppkes wrote:
I have also removed some superfluous checks for the format/fourcc at the
very start of the private data for A_QUICKTIME and V_QUICKTIME. This
shouldn't happen, since the private data always starts with the sample
description size, *then* the format/fourcc, in a Matroska file.
Don't remove any existing checks, they usually get added based on actual files.
"Should not" does not mean "does not".
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
I'm still a bit afraid about the following construct for both
A_QUICKTIME and V_QUICKTIME. It could lead to false positives. The fact
is that the format/fourcc ALWAYS starts at offset 4 of the private data
in a Matroska file. Does anyone sit on a file where it does not?
if (ff_codec_get_id(ff_codec_movaudio_tags,
AV_RL32(track->codec_priv.data))) {
fourcc = AV_RL32(track->codec_priv.data);
codec_id = ff_codec_get_id(ff_codec_movaudio_tags, fourcc);
}
Mats
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
