Webrtc's public video decoder api is defined by the interface
webrtc::VideoDecoder, declared in api/video_codecs/video_decoder.h.
The Decode method of this interface includes an argument of type
RTPFragmentationHeader*. There's also a corresponding argument in the
objc api, for the decode:...-method in the RTCVideoDecoder protocol,
declared in sdk/objc/Framework/Headers/WebRTC/RTCVideoCodec.h.
The RTPFragmentationHeader is not useful at all for video decoding. We
plan to delete that argument, replacing current api methods with new
ones with different signature/selector.
The first step is cl
https://webrtc-review.googlesource.com/c/src/+/72442, deprecating the
objc method and introducing a new one. This will be landed shortly.
I will update this thread with two-week advance warning with
transition instructions when the old methods in the objc and C++ api
are about to be deleted.
These changes affects applications that inject their own video
decoders into WebRTC. We expect the update to application code to be
easy: simply delete the unused argument.
Their are no immediate plans to delete RTPFragmentationHeader on the
encoder side, where it is passed to
webrtc::EncodedImageCallback::OnEncodedImage(), even though a cleanup
of related interfaces is desirable.
Regards,
/Niels Möller