Bug 46945 – Make suspendAnimations/resumeAnimations and setCSSAnimations traverse through subframes and remember state

• Status: RESOLVED
• Resolution: FIXED
• Component: Layout and Rendering

[Chris Marrin]

Currently suspendAnimations/resumeAnimations in WebCore take a document, but the pass through from WebKit, which allows access from LayoutTestController simply passes in the top-level document. The WebKit versions should traverse all subframes. Also, we don't remember the suspend state in AnimationController, so starting a new animation while suspended doesn't suspend.

I'm fixing these together because the test case I am writing needs the functionality in the former to test the latter.

[Simon Fraser (smfr)]

Dup of bug 46429?

[Chris Marrin]

*** Bug 46944 has been marked as a duplicate of this bug. ***

[Chris Marrin]

*** Bug 46429 has been marked as a duplicate of this bug. ***

[Chris Marrin]

<rdar://problem/8472265>

[Chris Marrin]

Created attachment 72995 [details]
Patch

[Simon Fraser (smfr)]

Comment on attachment 72995 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=72995&action=review

> LayoutTests/animations/animation-test-helpers.js:82
> +        }
> +    }

inIframe is redundant, because you could just test if (iframeId)...

> LayoutTests/animations/stop-animation-on-suspend.html:56
> +        setTimeout(suspend, 250);
> +        setTimeout(resume, 500);

Can we make this test faster?

[Darin Adler]

Comment on attachment 72995 [details]
Patch

We should not be adding new functions to Frame. We’ve been working hard slowly moving functions off of it and adding new ones is not a good idea. These function should go on the animation controller.

[Chris Marrin]

(In reply to comment #7)
> (From update of attachment 72995 [details])
> We should not be adding new functions to Frame. We’ve been working hard slowly moving functions off of it and adding new ones is not a good idea. These function should go on the animation controller.

Sorry, but I already checked in before seeing this. I have opened a new bug (https://bugs.webkit.org/show_bug.cgi?id=49073) to make this change.

[Chris Marrin]

Landed in http://trac.webkit.org/changeset/71424