[Webkit-unassigned] [Bug 53564] canHaveChildrenForEditing should be more efficient

Wed Feb 2 12:49:07 PST 2011

https://bugs.webkit.org/show_bug.cgi?id=53564

--- Comment #4 from Ryosuke Niwa <rniwa at webkit.org>  2011-02-02 12:49:07 PST ---
(In reply to comment #3)
> Qualified name comparison is just pointer comparison so we have to make sure we have code that takes advantage of that.
>
> I am not sure that HashSet<QualifiedName> is sufficiently specialized to do pointer hashing. If it was we might see faster performance. A way to work around that and test that assumption would be to use a HashSet<QualifiedNameImpl*>.

Ah, QualifiedNameImpl* seems to be considerably faster.  It's now 12s-13s, which is comparable to the performance obtained by making it a virtual function of Node.  But I found a number of bugs in this function (e.g. returns true for applet & embed elements) which could have been avoided if this function was virtual so I'm more inclined to make it a Node's virtual member function.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.