Treat a stopped AudioContext as closed.

Treat a stopped AudioContext as closed.

An AudioContext that ends up being detached&stopped by its
execution context is now treated on equal terms to one that
has been explicitly close()d from script. AudioContext methods
and operations over a context in such a state will now throw.

This restriction applies to OfflineAudioContext as well as
AudioContext.

R=rtoy
BUG=435867
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=194666

[sof, 2015-04-29 12:26:11]

sigbjornf@opera.com changed reviewers:
+ rtoy@chromium.org, tkent@chromium.org

[sof, 2015-04-29 12:26:11]

Please take a look.

Using AudioContexts in a stopped state leads to trouble here & there; try to
head off such trouble here.

[Raymond Toy, 2015-04-29 15:54:42]

Doesn't the fact that the offline context has been created keep the execution
context alive until the offline context is done?  Or does explicitly closing the
window for the execution context to go away?

https://codereview.chromium.org/1111213002/diff/30001/Source/modules/webaudio...
File Source/modules/webaudio/OfflineAudioContext.cpp (right):

https://codereview.chromium.org/1111213002/diff/30001/Source/modules/webaudio...
Source/modules/webaudio/OfflineAudioContext.cpp:102: ScriptPromise
OfflineAudioContext::startOfflineRendering(ScriptState* scriptState,
ExceptionState& exceptionState)
Why is exceptionState needed here? It doesn't seem to be used in this function.

[sof, 2015-04-29 16:05:22]

On 2015/04/29 15:54:42, Raymond Toy wrote:
> Doesn't the fact that the offline context has been created keep the execution
> context alive until the offline context is done?  Or does explicitly closing
the
> window for the execution context to go away?
> 

Yes; it cannot keep the document alive & prevent it from being finalized.

>
https://codereview.chromium.org/1111213002/diff/30001/Source/modules/webaudio...
> File Source/modules/webaudio/OfflineAudioContext.cpp (right):
> 
>
https://codereview.chromium.org/1111213002/diff/30001/Source/modules/webaudio...
> Source/modules/webaudio/OfflineAudioContext.cpp:102: ScriptPromise
> OfflineAudioContext::startOfflineRendering(ScriptState* scriptState,
> ExceptionState& exceptionState)
> Why is exceptionState needed here? It doesn't seem to be used in this
function.

Thanks, detritus from the initial change here -- will remove.

[Raymond Toy, 2015-04-29 16:07:29]

On 2015/04/29 16:05:22, sof wrote:
> On 2015/04/29 15:54:42, Raymond Toy wrote:
> > Doesn't the fact that the offline context has been created keep the
execution
> > context alive until the offline context is done?  Or does explicitly closing
> the
> > window for the execution context to go away?
> > 
> 
> Yes; it cannot keep the document alive & prevent it from being finalized.

Thanks.  I'm glad you understood my terrible typing skills. :-)
> 
> >
>
https://codereview.chromium.org/1111213002/diff/30001/Source/modules/webaudio...
> > File Source/modules/webaudio/OfflineAudioContext.cpp (right):
> > 
> >
>
https://codereview.chromium.org/1111213002/diff/30001/Source/modules/webaudio...
> > Source/modules/webaudio/OfflineAudioContext.cpp:102: ScriptPromise
> > OfflineAudioContext::startOfflineRendering(ScriptState* scriptState,
> > ExceptionState& exceptionState)
> > Why is exceptionState needed here? It doesn't seem to be used in this
> function.
> 
> Thanks, detritus from the initial change here -- will remove.

[Raymond Toy, 2015-04-29 16:07:50]

lgtm after you fix the nit.

[sof, 2015-04-29 16:16:56]

On 2015/04/29 16:07:50, Raymond Toy wrote:
> lgtm after you fix the nit.

done.

[sof, 2015-04-29 16:30:30]

The CQ bit was checked by sigbjornf@opera.com

[sof, 2015-04-29 16:30:31]

The patchset sent to the CQ was uploaded after l-g-t-m from rtoy@chromium.org
Link to the patchset: https://codereview.chromium.org/1111213002/#ps50001
(title: "startRendering() doesn't (directly) raise an exception")

[commit-bot: I haz the power, 2015-04-29 16:30:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1111213002/50001

[commit-bot: I haz the power, 2015-04-29 17:25:53]

Committed patchset #4 (id:50001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=194666