create one WebAudioSource for each CreateMediaStreamSource() call.

Create one WebAudioSource for each CreateMediaStreamSource() call.

Previously, the media stream track will create one source provider and let WebAudio use it. But this does not work when there are more than 1 WebAudio AudioContexts connecting to the same media stream.
This Cl fixes the problem by creating one source provider for each CreateMediaStreamSource() call.

NOTRY=true

BUG=354468
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170063

[no longer working on chromium, 2014-03-25 18:14:41]

Hi Raymond and/or Chris, could you please review this CL?
This Cl will create one AudioSourceProvider for each createMediaStreamSource(),
and MediaStreamSourceNode will own the AudioSourceProvider that is created by
the media stream track.

And this fixes the crash when there are multiple WebAudio AudioContexts
connecting to the same media stream.

Tommy, please take a look as well.

Thanks,
SX

[no longer working on chromium, 2014-03-25 22:24:22]

A gentle ping. Anyone in MTV can take an initial look? I would like to land it
tomorrow and try to get some test time for our testers.

Thanks,
SX

[Raymond Toy, 2014-03-25 22:35:41]

lgtm if it's true that the two versions give the same audioTrack.

https://codereview.chromium.org/202923007/diff/30001/Source/modules/webaudio/...
File Source/modules/webaudio/AudioContext.cpp (right):

https://codereview.chromium.org/202923007/diff/30001/Source/modules/webaudio/...
Source/modules/webaudio/AudioContext.cpp:406: OwnPtr<AudioSourceProvider>
provider = audioTrack->createWebAudioSource();
Are lines 405-406 here the same as the original lines 402-408? The original
finds the first entry in audioTracks[] that has an audioSourceProvider, but the
new code just uses audioTracks[0]. Are these the same?

[Tommy Widenflycht, 2014-03-26 13:15:35]

lgtm

[no longer working on chromium, 2014-03-26 13:31:11]

Thanks guys, landing soon.

https://codereview.chromium.org/202923007/diff/30001/Source/modules/webaudio/...
File Source/modules/webaudio/AudioContext.cpp (right):

https://codereview.chromium.org/202923007/diff/30001/Source/modules/webaudio/...
Source/modules/webaudio/AudioContext.cpp:406: OwnPtr<AudioSourceProvider>
provider = audioTrack->createWebAudioSource();
On 2014/03/25 22:35:41, Raymond Toy wrote:
> Are lines 405-406 here the same as the original lines 402-408? The original
> finds the first entry in audioTracks[] that has an audioSourceProvider, but
the
> new code just uses audioTracks[0]. Are these the same?

Yes, they are the same.
Chrome guarantees that the all the local track will have a source provider. So
the previous code just uses the source provider in the first track of the
stream. And there will be no source provider on the remote track.

The new code will use the first track to create a source provider.
For the local audio track, we create the source provider; but for remote audio
track, it will be NULL since we don't support it yet.

If there is only on CreateMediaStreamSource(), the behaviour will be exactly the
same as before; if there are multiple, the new code will work while the old code
won't.

[no longer working on chromium, 2014-03-26 14:35:30]

The CQ bit was checked by xians@chromium.org

[commit-bot: I haz the power, 2014-03-26 14:35:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/xians@chromium.org/202923007/30001

[commit-bot: I haz the power, 2014-03-26 14:35:57]

Change committed as 170063

[Raymond Toy, 2014-03-26 17:04:46]

https://codereview.chromium.org/202923007/diff/30001/Source/modules/webaudio/...
File Source/modules/webaudio/AudioContext.cpp (right):

https://codereview.chromium.org/202923007/diff/30001/Source/modules/webaudio/...
Source/modules/webaudio/AudioContext.cpp:406: OwnPtr<AudioSourceProvider>
provider = audioTrack->createWebAudioSource();
On 2014/03/26 13:31:11, xians1 wrote:
> On 2014/03/25 22:35:41, Raymond Toy wrote:
> > Are lines 405-406 here the same as the original lines 402-408? The original
> > finds the first entry in audioTracks[] that has an audioSourceProvider, but
> the
> > new code just uses audioTracks[0]. Are these the same?
> 
> Yes, they are the same.
> Chrome guarantees that the all the local track will have a source provider. So
> the previous code just uses the source provider in the first track of the
> stream. And there will be no source provider on the remote track.
> 
> The new code will use the first track to create a source provider.
> For the local audio track, we create the source provider; but for remote audio
> track, it will be NULL since we don't support it yet.
> 
> If there is only on CreateMediaStreamSource(), the behaviour will be exactly
the
> same as before; if there are multiple, the new code will work while the old
code
> won't.

Thanks for the very detailed explanation. This sounds excellent.