[Webkit-unassigned] [Bug 22672] ASSERT(m_table) when xhr.onabort creates another xhr or calls setTimeout

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=22672





------- Comment #2 from dimich at chromium.org  2008-12-04 23:34 PDT -------
There could be fast and right solutions to this, I think. The fast one is to
not use iterator in stopActiveDOMObjects but rather something like:
while(foo = activeObjects.begin()) {
foo->first->stop();
}
This can be an infinite loop so some counter and a limit could be used. Very
hacky since it does not guarantee that objects will be stopped.

Better solution could include a way to deny a newly created object from
becoming 'active'. For example, once SEC::stopActiveDOMObjects started to
terminate active ones, external code can create XHR and call send() but xhr
will be zombied from the start and not cause any listener callbacks. That feels
right because if there is a concept of Active Object, the sequence of gaining
'active' status should be spelled out just as a sequence of loosing such
status.


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