W3C home > Mailing lists > Public > public-webrtc@w3.org > May 2015

Obj-C RTCMediaStreamTrackDelegate reports "ended" when it shouldn't

From: Iñaki Baz Castillo <ibc@aliax.net>
Date: Mon, 25 May 2015 11:02:08 +0200
Message-ID: <CALiegf=9hZeqRMdcG3CG8iBLn85-ddNZAfKff9jvm2r=FMjtvg@mail.gmail.com>
To: "public-webrtc@w3.org" <public-webrtc@w3.org>
Scenario:

- iOS device generates an audio/video offer and establishes a session
with a remote peer (let's say Chrome desktop).

- Later, Chrome puts the call "on hold" by sending its previous SDP
with all the m= lines having "a=sendonly" (but still same msid and so
on, the SDP is just mangled at JS level).

- iOS receives the re-offer, calls to setRemoteDescription() followed
by createAnswer() and setLocalDescription().

- The remote RTCMediaStreamTracks of the RTCPeerConnection report
RTCTrackStateEnded state (by means of their
RTCMediaStreamTrackDelegate instances).

Why? Does it make sense? Those are the exact same MediaStreamTracks
than before. It is just that now they are offered with "a=sendonly"
which means that the remote peer just wants to send media to iOS but
not receive from it. This is: the tracks remain alive! why is
RTCTrackStateEnded reported?

Thanks a lot.

-- 
Iñaki Baz Castillo
<ibc@aliax.net>
Received on Monday, 25 May 2015 09:02:58 UTC

This archive was generated by hypermail 2.3.1 : Monday, 23 October 2017 15:19:44 UTC