Removing PropertyWrapper makes the AnimationBase simpler.
Created attachment 139034 [details] Patch Proposed patch.
With this change will be easier to make gPropertyWrappers ref counted. It is the next step.
(In reply to comment #2) > With this change will be easier to make gPropertyWrappers ref counted. It is the next step. Why do you want to do this?
Right now gPropertyWrappers is not deleted, we need to find a way to delete it. Instead of be ref counted we can just create a destroy method in the AnimationPropertyHandler and call it when ~AnimationController is called. (In reply to comment #3) > (In reply to comment #2) > > With this change will be easier to make gPropertyWrappers ref counted. It is the next step. > > Why do you want to do this?
Why does it have to be deletable?
Created attachment 139039 [details] Patch Updated patch
As discussed in the IRC, does not make sense we make gPropertyWrappers deletable. However, this patch fixes the layer violation about AnimationController accessing AnimationBase property handlers and make AnimationBase simpler. (In reply to comment #5) > Why does it have to be deletable?
Created attachment 139042 [details] Patch Proposed patch.
Comment on attachment 139042 [details] Patch Attachment 139042 [details] did not pass qt-wk2-ews (qt): Output: http://queues.webkit.org/results/12547004
Comment on attachment 139042 [details] Patch Attachment 139042 [details] did not pass mac-ews (mac): Output: http://queues.webkit.org/results/12523955
Comment on attachment 139042 [details] Patch Attachment 139042 [details] did not pass gtk-ews (gtk): Output: http://queues.webkit.org/results/12528956
Comment on attachment 139042 [details] Patch Clear flags. A new patch is being cooked.
Comment on attachment 139042 [details] Patch Attachment 139042 [details] did not pass win-ews (win): Output: http://queues.webkit.org/results/12490898
Comment on attachment 139042 [details] Patch Attachment 139042 [details] did not pass qt-ews (qt): Output: http://queues.webkit.org/results/12528958
Comment on attachment 139042 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=139042&action=review > Source/WebCore/ChangeLog:8 > + of property handlers. This patch remove the property handlers for a > + separated class making AnimationBase simpler. I'd use the term "move" rather than "remove". > Source/WebCore/page/animation/AnimationPropertyHandler.h:41 > +class AnimationPropertyHandler { I think PropertyAnimation would be a better name for this class, or maybe even CSSPropertyAnimation.
Created attachment 139108 [details] Patch Updated patch.
Created attachment 139112 [details] Patch Updated patch. Fix changelog.
Comment on attachment 139112 [details] Patch Attachment 139112 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/12551293 New failing tests: compositing/overflow-trumps-transform-style.html transforms/3d/general/3dtransform-values.html
Created attachment 139251 [details] Archive of layout-test-results from ec2-cr-linux-03 The attached test failures were seen while running run-webkit-tests on the chromium-ews. Bot: ec2-cr-linux-03 Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'> Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick
Created attachment 139272 [details] Patch Fix crash in the previous patch.
Created attachment 139277 [details] Patch Updated patch.
Comment on attachment 139277 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=139277&action=review As long is this is just moving code to CSSPropertyAnimation, then I'm ok with it. > Source/WebCore/ChangeLog:10 > + property handlers. This patch moves the property handlers for a separated > + class making AnimationBase simpler. handlers *to a separate* class
Created attachment 139366 [details] Patch final patch.
Comment on attachment 139366 [details] Patch Clearing flags on attachment: 139366 Committed r115581: <http://trac.webkit.org/changeset/115581>