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

Re: [webrtc-pc] Effect of a BYE on RtpReceiver.track

From: Harald Alvestrand via GitHub <sysbot+gh@w3.org>
Date: Wed, 18 Jan 2017 14:37:13 +0000
To: public-webrtc-logs@w3.org
Message-ID: <issue_comment.created-273491276-1484750231-sysbot+gh@w3.org>
we want BYE to be sent if we're never going to use that SSRC again. So
 flipping .enabled shouldn't touch it. .ended is set from .stop, but 
we might want to resurrect it even then, so what do we do when

pc.addTrack(track1) => sender1
track1.stop()
sender1.replaceTrack(track2)

If the sequence

pc.addTrack(track1) => sender1
sender1.replaceTrack(track2)

continues sending on the same ssrc, it seems that the first sequence 
should also do so, and thus we shouldn't be sending BYE on ended.

after a transceiver.stop() we're not going to use that SSRC again, so 
sending a bye then seems OK.
similar logic brings me to think that a BYE says "I'm not going to be 
talking to you on this ssrc again", so doing track.stop and firing an 
ended event seems sensible.

If we have examples in the wild of people doing BYE & then 
resurrecting the SSRC, this logic falls flat.


-- 
GitHub Notification of comment by alvestrand
Please view or discuss this issue at 
https://github.com/w3c/webrtc-pc/issues/961#issuecomment-273491276 
using your GitHub account
Received on Wednesday, 18 January 2017 14:37:24 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 4 June 2019 15:32:41 UTC