RESOLVED FIXED 74692
invalidateNodeListsCacheAfterAttributeChanged has too many callers
https://bugs.webkit.org/show_bug.cgi?id=74692
Summary invalidateNodeListsCacheAfterAttributeChanged has too many callers
Ryosuke Niwa
Reported 2011-12-16 00:20:48 PST
We shouldn't have to call invalidateNodeListsCacheAfterAttributeChanged in various places in DOM code.
Attachments
cleanup and fixes a bug (17.20 KB, patch)
2011-12-16 01:02 PST, Ryosuke Niwa
no flags
Fixed if for MICRODATA (17.20 KB, patch)
2011-12-16 01:06 PST, Ryosuke Niwa
sam: review+
Ryosuke Niwa
Comment 1 2011-12-16 01:02:55 PST
Created attachment 119581 [details] cleanup and fixes a bug
Ryosuke Niwa
Comment 2 2011-12-16 01:06:46 PST
Created attachment 119582 [details] Fixed if for MICRODATA
Ryosuke Niwa
Comment 3 2011-12-16 13:37:05 PST
Ping reviewers.
Sam Weinig
Comment 4 2011-12-16 14:26:58 PST
Comment on attachment 119582 [details] Fixed if for MICRODATA View in context: https://bugs.webkit.org/attachment.cgi?id=119582&action=review > Source/WebCore/dom/Node.cpp:1038 > + if (attrName != classAttr > +#if ENABLE(MICRODATA) > + && attrName != itemscopeAttr > + && attrName != itempropAttr > +#endif > + && attrName != nameAttr) > + return; If this needs to be in sync with something, can we extract this logic into a shared function?
Ryosuke Niwa
Comment 5 2011-12-16 14:31:40 PST
Comment on attachment 119582 [details] Fixed if for MICRODATA View in context: https://bugs.webkit.org/attachment.cgi?id=119582&action=review Thanks for the review! >> Source/WebCore/dom/Node.cpp:1038 >> + return; > > If this needs to be in sync with something, can we extract this logic into a shared function? It's hard to do at the moment because information is hidden in NodeListsNodeData but it'll be straight-forward once we turn NodeListsNodeData into a per-document cache (which I'm planning to do on the next patch).
Ryosuke Niwa
Comment 6 2011-12-16 15:15:15 PST
Note You need to log in before you can comment on or make changes to this bug.