[FFmpeg-devel] [PATCH v3 3/6] avformat/s337m: Make available as subdemuxer

[Gaullier Nicolas]

>Look at spdif in wav.
Indeed, I missed that, spdif and s337m are very similar technically, and it means I have to rework my patch for the integration of s337m in wavdec, but I still do not catch why another demuxer should be created : spdifdec.c is actually quite similar to the existing s337m.c
Perhaps you find unacceptable only the use of the avpriv_ prefix instead of ff_ ? I understand, but in a previous cover letter, I mentioned that the idea was to make this available to build a filter that would decode s337m in a filter graph; but yes, maybe I should stay with ff_ for now and migrate to avpriv_ later when submitting the new filter, that would make sense, moreover the prototype may be changed too depending on how the implementation of the filter turns to be (and I have no experience in writing a filter for ffmpeg for now, so...).

In my opinion, here are the most relevant differences between spdif and s337m, it is a matter of use cases :
- s337m may be split in two mono streams and may require an "audio merge" to be decodable, which suggests the use of a filter graph and a dedicated filter to decode s337/dolby_e (so the use of public methods for a dedicated format, indeed)
- dolby E is a codec one either don't want to see (ie. pass through) or one want to get rid off/decode to uncompressed/pcm, whereas ac3 for example is a very popular codec one usually wants to stream copy: so an audio filter to decode s337/dolbyE seems interesting whereas an audio filter to decode spdif/ac3 would be absolutely useless
- s337m may be frame wrapped (ex: mxf) (probing method differs)

At the end, in my current understanding, here is what I should correct :
- make the integration of s337m in wavdec similar to that of spdif
- replace the avpriv_* by ff_* ... for now

Anyway, I will wait your feedback and until we completely agree before going anywhere further.

Nicolas
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel at ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".