On Mon, Feb 15, 2016 at 11:39 AM, wm4 <nfx...@googlemail.com> wrote: > On Mon, 15 Feb 2016 10:21:19 +0100 > Steve Lhomme <rob...@videolabs.io> wrote: > >> On Mon, Feb 15, 2016 at 10:16 AM, Carl Eugen Hoyos <ceho...@ag.or.at> wrote: >> > Steve Lhomme <robux4 <at> videolabs.io> writes: >> > >> >> I implemented the same change in VLC. >> > >> > From a quick look, you added code to the decoder >> > wrapper but how is remuxing supposed to work? >> >> In VLC ? There's no Matroska muxer so it would use the ffmpeg one. >> >> >> But I think it would be better if the decoder works >> >> with the lighter version of the extra data. >> > >> > Why? >> > I mean where would the incomplete extra-data come from? >> >> Look at the code in ffmpeg I liked. It's just header boilerplate. > > Matroska requires a lot of this, so where's the problem?
It's not a problem. I just thought making the code more flexible on the decoder could help all parts of the code and keep the hacks in one place. The one piece of the code that actually uses the data. But then the header would still need to be reconstructed in the MOV muxer when the data are missing. > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
