Bug 224430 – Add support for RTCEncodedFrame metadata

• Status: RESOLVED
• Resolution: FIXED
• Component: WebRTC

[youenn fablet]

Add support for RTCEncodedFrame metadata

[youenn fablet]

Created attachment 425735 [details]
Patch

[youenn fablet]

Created attachment 425737 [details]
Patch

[Eric Carlson]

Comment on attachment 425737 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=425737&action=review

> Source/WebCore/Modules/mediastream/libwebrtc/LibWebRTCRtpTransformableFrame.cpp:114
> +    return { frameId, WTFMove(dependencies), metadata.GetWidth(), metadata.GetHeight(), metadata.GetSpatialIndex(), metadata.GetTemporalIndex(), m_rtcFrame->GetSsrc(), { } };

Will we eventually support contributingSources? If so, it would be good to have a bug and reference it here.

[youenn fablet]

(In reply to Eric Carlson from comment #3)
> Comment on attachment 425737 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=425737&action=review
> 
> > Source/WebCore/Modules/mediastream/libwebrtc/LibWebRTCRtpTransformableFrame.cpp:114
> > +    return { frameId, WTFMove(dependencies), metadata.GetWidth(), metadata.GetHeight(), metadata.GetSpatialIndex(), metadata.GetTemporalIndex(), m_rtcFrame->GetSsrc(), { } };
> 
> Will we eventually support contributingSources? If so, it would be good to
> have a bug and reference it here.

contributing sources is mostly for audio mixers where a server is mixing several audio in one stream but web page might still want to expose who is actually speaking.

I will file a bug to track it though since CSRC is not audio specific.

[EWS]

Committed r275830 (236400@main): <https://commits.webkit.org/236400@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 425737 [details].

[Radar WebKit Bug Importer]

<rdar://problem/76549361>

[youenn fablet]

> contributing sources is mostly for audio mixers where a server is mixing
> several audio in one stream but web page might still want to expose who is
> actually speaking.
> 
> I will file a bug to track it though since CSRC is not audio specific.

Let's make the attribute undefined if there are no CSRCs so that we do not surface it until we have something.