1199098 - Move audio/video rendering and A/V sync out of MDSM

Status: NEW

(Core :: Audio/Video: Playback, defect, P3)

Description

[JW Wang [:jwwang] [:jw_wang]]

This is required by TV projects and CDM which want to do rendering using specific hardware and manage A/V sync on their own.

Comment 1

[JW Wang [:jwwang] [:jw_wang]]

Here is the plan:

1. Create an abstract class MediaSink to encapsulate the idea of audio/video rendering and manage A/V sync.

2. Create a subclass of MediaSink to wrap DecodedAudioDataSink. MDSM will still do video rendering on its own and sync video to the position reported by this MediaSink.

3. Create a subclass of MediaSink to wrap DecodedStream. MDSM will still do video rendering on its own and sync video to the position reported by this MediaSink. However, MDSM will no longer have to care about whether audio is captured or not.

4. Create a subclass of MediaSink to render both audio and video. MDSM will no longer render video but only update playback position according to the position reported by this MediaSink.

5. Create a subclass of MediaSink to render both audio and video using specific hardware per TV project's requirement.

6. Create a subclass of MediaSink to render encrypted content per CDM's requirement.

Comment 2

[Anthony Jones (:ajones, :kentuckyfriedtakahe, :k17e)]

Mass change P2 -> P3