On Thu, May 30, 2019 at 08:33:35PM +0200, Reimar Döffinger wrote: > > > On 29.05.2019, at 19:51, Swaraj Hota <swarajhota...@gmail.com> wrote: > >> > >> Entirely depends on the purpose. > >> If the aim is to have a working demuxer, I think it is acceptable from > >> what I looked at. > >> I guess the only concern might be that not supporting piped input would be > >> a bit of a regression from the first patch. > > > > Is there another simpler way to support piped input? Can you explain a > > bit? > > The mov demuxer code isn't really all that complicated. > The basic idea is: if the input is not seekable, decided whether to read > audio or video based on which one has the lowest file position (a bit like > your original patch, but using it only for the "audio or video next" decision. > If input is seekable, pick the stream with the lowest next timestamp as your > patch does now. > It should be not that complicated in principle, but the details can make it a > pain sometimes.
Okay, I think I get the basic idea now. I will surely try to implement this later in another patch. Thanx a lot! Swaraj _______________________________________________ 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".
