54025 2011-02-08 11:43:15 -0800 Adjusting relatedTarget can be much faster and more accurate 2012-01-12 14:42:27 -0800 1 1 1 Unclassified WebKit DOM 528+ (Nightly build) PC OS X 10.5 NEW P2 Normal --- 52065 52963 1 dglazkov dglazkov dglazkov eric hayato rolandsteiner tonikitoo oldest_to_newest 347571 0 dglazkov 2011-02-08 11:43:15 -0800 Once we have O(1) access to shadowRoot, we can: 1) Easily determine whether we even need to walk anything. 2) Use scopes to check whether we need to adjust relatedTarget. See http://dev.w3.org/2006/xbl2/#the-mouseover-and-mouseout-events for more details. 3) Use scope chain to walk more quickly. 453674 1 hayato 2011-08-19 00:03:25 -0700 I think we already have O(1) access to shadowRoot. Let me rewrite adjust relatedTarget code. 453822 2 dglazkov 2011-08-19 07:50:08 -0700 (In reply to comment #1) > I think we already have O(1) access to shadowRoot. Let me rewrite adjust relatedTarget code. SVG is still blocking you :(. 454426 3 hayato 2011-08-21 16:47:39 -0700 Thank you for the info. SVG still looks blocking as you said. :( (In reply to comment #2) > (In reply to comment #1) > > I think we already have O(1) access to shadowRoot. Let me rewrite adjust relatedTarget code. > > SVG is still blocking you :(. 535180 4 eric 2012-01-12 14:38:11 -0800 This isn't causing: http://samples.msdn.microsoft.com/ietestcenter/domevents/mouseleave.relatedTarget.html and: http://samples.msdn.microsoft.com/ietestcenter/domevents/mouseenter.ctrlKey.html is it? 535182 5 eric 2012-01-12 14:38:53 -0800 Sorry, http://samples.msdn.microsoft.com/ietestcenter/domevents/mouseenter.relatedTarget.html 535187 6 eric 2012-01-12 14:42:27 -0800 Nevermind. We don't support those two events. https://bugs.webkit.org/show_bug.cgi?id=18930