Fix crash when registering abs-send-time to AudioSend/ReceiveStream.

Fix crash when registering abs-send-time to AudioSend/ReceiveStream.

Introduced with r14870.

BUG=b/32591921

Committed: https://crrev.com/572ae1212b488d874f6afcf87bc536319b122df7
Cr-Commit-Position: refs/heads/master@{#14883}

[stefan-webrtc, 2016-11-02 07:33:39]

stefan@webrtc.org changed reviewers:
+ solenberg@webrtc.org

[stefan-webrtc, 2016-11-02 07:33:40]

This solves the issue, but I'm not sure we should have an NOTREACHED() on header
extensions which we don't know about. The application can try to register any
extension by changing the SDP.

WDYT? Should I move the LOG to the else {} instead?

[stefan-webrtc, 2016-11-02 09:42:45]

stefan@webrtc.org changed reviewers:
+ tina.legrand@webrtc.org

[stefan-webrtc, 2016-11-02 09:42:46]

+tina

[tlegrand-webrtc, 2016-11-02 09:46:28]

lgtm

[stefan-webrtc, 2016-11-02 09:46:44]

The CQ bit was checked by stefan@webrtc.org

[commit-bot: I haz the power, 2016-11-02 09:46:50]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2016-11-02 10:10:13]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-11-02 10:10:15]

Description was changed from

==========
Fix crash when registering abs-send-time to AudioSend/ReceiveStream.

Introduced with r14870.

BUG=b/32591921
==========

to

==========
Fix crash when registering abs-send-time to AudioSend/ReceiveStream.

Introduced with r14870.

BUG=b/32591921

Committed: https://crrev.com/572ae1212b488d874f6afcf87bc536319b122df7
Cr-Commit-Position: refs/heads/master@{#14883}
==========

[commit-bot: I haz the power, 2016-11-02 10:10:16]

Patchset 1 (id:??) landed as
https://crrev.com/572ae1212b488d874f6afcf87bc536319b122df7
Cr-Commit-Position: refs/heads/master@{#14883}

[the sun, 2016-11-07 09:12:29]

On 2016/11/02 07:33:40, stefan-webrtc (holmer) wrote:
> This solves the issue, but I'm not sure we should have an NOTREACHED() on
header
> extensions which we don't know about. The application can try to register any
> extension by changing the SDP.
> 
> WDYT? Should I move the LOG to the else {} instead?

The reasoning is that the outer layer (WVoE/MC) should have already verified the
SDP so that here we will always be called with valid data. 

webrtc::RtpExtension::IsSupportedForAudio() is used to filter extensions in
WVoMC::Set[Send|Recv]Parameters(). If we don't support AST for audio, the
function should be updated and you can drop the logging here. Can you create a
CL for this?