ElementAttributeData should be dumber.
Created attachment 173958 [details] Scandinavian patch
Comment on attachment 173958 [details] Scandinavian patch Attachment 173958 [details] did not pass qt-ews (qt): Output: http://queues.webkit.org/results/14826284
Comment on attachment 173958 [details] Scandinavian patch Attachment 173958 [details] did not pass qt-wk2-ews (qt): Output: http://queues.webkit.org/results/14832025
Created attachment 173959 [details] Meh patch
Comment on attachment 173959 [details] Meh patch View in context: https://bugs.webkit.org/attachment.cgi?id=173959&action=review > Source/WebCore/dom/StyledElement.cpp:146 > + // For some reason, we allow CSSOM wrappers for inline style declarations to outlive the element. > + if (PropertySetCSSStyleDeclaration* cssomWrapper = inlineStyleCSSOMWrapper()) > + cssomWrapper->clearParentElement(); The comment doesn't seem that helpful. Do you mean that the wrapper should keep the element alive? > Source/WebCore/dom/StyledElement.cpp:151 > - return ensureInlineStyle()->ensureInlineCSSStyleDeclaration(this); > + return ensureMutableInlineStyle()->ensureInlineCSSStyleDeclaration(this); Would be good to move the wrapper ownership to StyledElement too at some point. The current propertySetCSSOMWrapperMap hackery is silly.
Committed r134539: <http://trac.webkit.org/changeset/134539>
This broke Windows http://build.webkit.org/builders/Apple%20Win%20Debug%20%28Build%29