Re: [mediacapture-transform] MSTP's VideoFrame lifetime management model assumes a single consumer (#56)

One way of addressing memory management concerns would be to have a consistent strategy with respect to implicit `close()` operations. 

The argument for MSTG implicitly calling `VideoFrame.close()` is that converting to a `MediaStreamTrack` is typically the last step in the pipeline.  However, if that's the logic then I'd expect that other rendering approaches (e.g. canvas or WebGL/GPU) would also include an implicit `close()`.  Is that always true??

WebCodecs `encode()` method does not implicitly call `close()`. If the `VideoFrame` is subsequently rendered (e.g. self view) an implicit `close()` might prevent a memory leak. But will this always be the case?  

-- 
GitHub Notification of comment by aboba
Please view or discuss this issue at https://github.com/w3c/mediacapture-transform/issues/56#issuecomment-891454577 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Tuesday, 3 August 2021 02:07:22 UTC