Follow up the recent HTMLSelectElement improvements with a little bit more
Created attachment 111075 [details] Patch
Comment on attachment 111075 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=111075&action=review The code looks much better after this patch. r=me provided assertions in toHTMLSelectElement are fixed. > Source/WebCore/html/HTMLSelectElement.cpp:471 > const Vector<HTMLElement*>& items = listItems(); This local variable seems unnecessary. > Source/WebCore/html/HTMLSelectElement.h:168 > + mutable Vector<HTMLElement*> m_listItems; I'm not sure making this mutable is necessarily cleaner. > Source/WebCore/html/HTMLSelectElement.h:181 > + mutable bool m_shouldRecalcListItems; Ditto. > Source/WebCore/html/HTMLSelectElement.h:200 > + ASSERT(!node || node->isHTMLElement()); This should be isHTMLSelectElement. > Source/WebCore/html/HTMLSelectElement.h:206 > + ASSERT(!node || node->isHTMLElement()); Ditto.
Comment on attachment 111075 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=111075&action=review >> Source/WebCore/html/HTMLSelectElement.h:168 >> + mutable Vector<HTMLElement*> m_listItems; > > I'm not sure making this mutable is necessarily cleaner. How so? This is exactly what mutable is for.
Committed r97533: <http://trac.webkit.org/changeset/97533>
Windows build fix attempt committed in http://trac.webkit.org/changeset/97534.
Thanks for fixing the chromium windows build in http://trac.webkit.org/changeset/97535.
GTK build fix landed in http://trac.webkit.org/changeset/97539.
Windows build is still broken ...
(In reply to comment #8) > Windows build is still broken ... Sorry for blaming you, http://trac.webkit.org/changeset/97536 is the culprit.