[webkit-reviews] review granted: [Bug 210167] Adopt interface AVAudioRoutingArbiter for Mac : [Attachment 396311] Patch (with rename enabled)

[bugzilla-daemon at webkit.org]

Eric Carlson <eric.carlson at apple.com> has granted Jer Noble
<jer.noble at apple.com>'s request for review:
Bug 210167: Adopt interface AVAudioRoutingArbiter for Mac
https://bugs.webkit.org/show_bug.cgi?id=210167

Attachment 396311: Patch (with rename enabled)

https://bugs.webkit.org/attachment.cgi?id=396311&action=review




--- Comment #14 from Eric Carlson <eric.carlson at apple.com> ---
Comment on attachment 396311
  --> https://bugs.webkit.org/attachment.cgi?id=396311
Patch (with rename enabled)

View in context: https://bugs.webkit.org/attachment.cgi?id=396311&action=review

>
Source/WebKit/UIProcess/Media/cocoa/AudioSessionRoutingArbitratorProxyCocoa.mm:
122
> +	       for (auto& callback : m_enqueuedCallbacks)
> +		   callback(error ? RoutingArbitrationError::Failed :
RoutingArbitrationError::None, defaultDeviceChanged ? DefaultRouteChanged::Yes
: DefaultRouteChanged::No);
> +
> +	       m_enqueuedCallbacks.clear();

Is there any chance that calling the callback could allow pending messages to
be processed? If so, it would be safer to copy the callbacks to a new vector
and clear m_enqueuedCallbacks first.