[webkit-reviews] review denied: [Bug 12251] REGRESSION (r18822-r18823): Assertion failure opening document with non-existent resources (dom/xhtml/level2/html/HTMLIFrameElement11.xhtml) : [Attachment 12424] Patch v1

bugzilla-request-daemon at macosforge.org bugzilla-request-daemon at macosforge.org
Sun Jan 14 06:03:10 PST 2007


Darin Adler <darin at apple.com> has denied Darin Adler <darin at apple.com>'s
request for review:
Bug 12251: REGRESSION (r18822-r18823): Assertion failure opening document with
non-existent resources (dom/xhtml/level2/html/HTMLIFrameElement11.xhtml)
http://bugs.webkit.org/show_bug.cgi?id=12251

Attachment 12424: Patch v1
http://bugs.webkit.org/attachment.cgi?id=12424&action=edit

------- Additional Comments from Darin Adler <darin at apple.com>
We definitely need to fix this! Can't have layout tests failing.

But HashMap remove already has the behavior of removing it only if it's there,
so no need for the contains call.

And if the object that comes back from _objectForIdentifier is nil, then the
caller needs to not send the notification.

I think we should land a fixed-up version of this now. The real problem is that
the loader is sending multiple "done loading" calls to the client for the same
resource -- it can be fixed entirely on the WebCore side. But it's a complex
job that we should do later. This approach is correct for now.



More information about the webkit-reviews mailing list