- From: <bugzilla@jessica.w3.org>
- Date: Fri, 17 Apr 2015 07:57:19 +0000
- To: public-webapps-bugzilla@w3.org
https://www.w3.org/Bugs/Public/show_bug.cgi?id=23887 --- Comment #180 from Hayato Ito <hayato@chromium.org> --- Bad news for you :( > Exclude shadow insertion points whose containing shadow root has older shadow roots from INSERTION-POINTS I'm afraid this this would take us to the lose situation. For example: (Let me allow to use the pseudo notation to describe a tree of trees) <div id="shadow-host"> <shadow-root id="oldest-shadow-root-of-shadow-host"> <div id="a"> <shadow-root id="youngest-shadow-root-of-shadow-host"> <div id="shadow-host-b"> <shadow id="shadow"> <shadow-root id="shadow-root-of-shadow-host-b"> <content id="content"> Note that #a's destination insertion points is [#shadow, #content]. If an event happens on #a, the event path should be (in terms of the new proposal): [#a => #oldest-shadow-root-of-shadow-host => #content => #shadow-root-of-shadow-host-b => #shadow => #shadow-host-b => #youngest-shadow-root-of-shadow-host => #shadow-host] But according to the algorithm of candidate 2, the event path would be: [#a => #content => #shadow-root-of-shadow-host-b => #shadow-host-b => #youngest-shadow-root-of-shadow-host => #shadow-host] The point is the proposed algorithm can't handle the case where re-distribution happens through shadow insertion points *and* content insertion points. I'll take another look later. -- You are receiving this mail because: You are the QA Contact for the bug.
Received on Friday, 17 April 2015 07:57:22 UTC