W3C home > Mailing lists > Public > public-media-capture-logs@w3.org > January 2017

Re: [mediacapture-image] is grabFrame necessary?

From: Harald Alvestrand via GitHub <sysbot+gh@w3.org>
Date: Thu, 19 Jan 2017 19:50:44 +0000
To: public-media-capture-logs@w3.org
Message-ID: <issue_comment.created-273879497-1484855443-sysbot+gh@w3.org>
.... except that the entity getting the frame is a consumer of the 
MediaStreamTrack, not the producer as it was in the canvas case, so 
the model doesn't quite fit. We want to grab frames from any sort of 
track, without having to have the frame-grabbing travel all the way to
 the producer having to know that this is a track of the type you can 
grab frames from.

Once upon a time we had separate IDL classes for VideoTrack and 
AudioTrack, but after a year of nobody finding any API we wanted to 
have different on the two, we went ahead and merged them. Separating 
them back out would be a pain in the posterior, I think.

To me it seems simpler to have grabFrame() be a method on the object 
you connect a MediaStreamTrack to, just the way it's currently 
specified.

But at the end of the day, it's all control surface. The real data 
travels by whatever means makes sense for the implementation.


-- 
GitHub Notification of comment by alvestrand
Please view or discuss this issue at 
https://github.com/w3c/mediacapture-image/issues/144#issuecomment-273879497
 using your GitHub account
Received on Thursday, 19 January 2017 19:50:52 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:27:31 UTC