Bug 100047 – Clear m_mediaSource reference when the MediaSource is closed

• Status: RESOLVED
• Resolution: FIXED
• Component: Media

[Aaron Colwell]

The existing code in HTMLMediaElement holds the reference to m_mediaSource longer then it needs to. The object shouldn't be used again after it has transitioned to "closed" so I'm adding code to clear the reference when this transition happens.

[Aaron Colwell]

Created attachment 170019 [details]
Patch

[Early Warning System Bot]

Comment on attachment 170019 [details]
Patch

Attachment 170019 [details] did not pass qt-wk2-ews (qt):
Output: http://queues.webkit.org/results/14489715

[Darin Adler]

Comment on attachment 170019 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=170019&action=review

> Source/WebCore/ChangeLog:12
> +        No new tests because the change isn't visible to JavaScript.

Understood that it is not “visible to JavaScript”. What end user detectable effect does it have?

[Aaron Colwell]

(In reply to comment #3)
> (From update of attachment 170019 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=170019&action=review
> 
> > Source/WebCore/ChangeLog:12
> > +        No new tests because the change isn't visible to JavaScript.
> 
> Understood that it is not “visible to JavaScript”. What end user detectable effect does it have?

I don't believe it has an end user detectable effect. This change simply makes the object state more consistent so code can't access the object when it isn't supposed to.

[Aaron Colwell]

Comment on attachment 170019 [details]
Patch

resetting commit-queue? since the qt-wk2 failure doesn't appear to be related to my change.

[WebKit Review Bot]

Comment on attachment 170019 [details]
Patch

Clearing flags on attachment: 170019

Committed r132280: <http://trac.webkit.org/changeset/132280>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.