On Thu, Nov 3, 2016 at 7:39 PM, Andreas Cadhalpun <andreas.cadhal...@googlemail.com> wrote: > On 03.11.2016 09:52, Hendrik Leppkes wrote: >> Hence my point about trying to stick to a common behavior scheme >> without looking at what *OUR* decoder needs - there could be others, >> after all. >> Its common to write extradata into codecpar if its present. You can >> try to dispute that as long as you wish, but that doesn't make it any >> less true. >> >> avcodec is not the only thing people use avformat with, and >> vice-versa, so sticking to common patterns makes the life easier for >> those. >> >> Do we really need to write this down somewhere to convince you, what >> do we have the ML for then if we can't inform contributors about >> common practice? > > What you need to convince me are good arguments. > The reason why I asked about documentation is that if it's not documented, > your argument about other decoders, that you repeated above, is void, > because others shouldn't rely on undocumented behavior, as our > documentation states clearly [1]: > "Behavior in undocumented situations may change slightly (and be documented)." > > So if you want to be able to rely on the behavior you describe, you'll > have to get it documented. > As you seem to care about this a lot, I don't understand why you appear > to be opposed to the idea of documenting it. > >
Since you want docs, I even found one extremely specific to this particular case: https://ffmpeg.org/ffmpeg-formats.html#apng "All headers, but the PNG signature, up to (but not including) the first fcTL chunk are transmitted as extradata." I'm sure you will argue that side-data is still extradata of some sort, but fact is the apng demuxer changed behavior recently, and the documentation suggests a certain content of extradata is to be expected. - Hendrik _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
