Bug 218748 – [GStreamer] refactor video encoder with WebKit style

• Component: WebRTC
• Status: RESOLVED FIXED

[Víctor M. Jáquez L.]

libwebrtc/GStreamerVideoEncoder is written in a GStreamer code-style. It would be better to refactor it using C++/WebKit style, as the other GStreamer elements inside WebKit.

[Víctor M. Jáquez L.]

Created attachment 413694 [details]
Patch

[Xabier Rodríguez Calvar]

Let's wait for Phil and Thibault's comments, since IIRC, it was their idea to keep it this way to ease a possible upstreaming process to GStreamer eventually.

[Thibault Saunier]

(In reply to Xabier Rodríguez Calvar from comment #2)
> Let's wait for Phil and Thibault's comments, since IIRC, it was their idea
> to keep it this way to ease a possible upstreaming process to GStreamer
> eventually.

It indeed was the plan and the reason why we used the GStreamer coding style. I never took the time to make it happen and am not 100% sure what other GStreamer maintainer will think of it so I have no hard opinion at that point.

[Philippe Normand]

Comment on attachment 413694 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=413694&action=review

> Source/WebCore/platform/mediastream/libwebrtc/GStreamerVideoEncoder.cpp:39
> +    GST_STATIC_PAD_TEMPLATE ("src", GST_PAD_SRC, GST_PAD_ALWAYS, GST_STATIC_CAPS ("video/x-h264;video/x-vp8;video/x-vp9"));

vp9 is added here, but I see no mention of a vp9enc below. Maybe add vp9 in a follow-up patch, if this patch is meant to be a refactoring only.

[Víctor M. Jáquez L.]

Created attachment 414799 [details]
Patch

[EWS]

Committed r270187: <https://trac.webkit.org/changeset/270187>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 414799 [details].

[Radar WebKit Bug Importer]

<rdar://problem/71761576>