[Webkit-unassigned] [Bug 12665] REGRESSION: Assertion failure in scheduleRelayout visiting Bookmarks view while a page is loading

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Sun Mar 11 12:33:30 PDT 2007


mitz at webkit.org changed:

           What    |Removed                     |Added
                 CC|                            |mitz at webkit.org

------- Comment #3 from mitz at webkit.org  2007-03-11 12:33 PDT -------
(In reply to comment #2)
> Looks to me like the issue is that when the bookmarks view is up there's also
> an empty document in the frame. When we return from the bookmarks view to the
> non-HTML view we "close" the document, which results in adding a body element,
> which results in a desire to do layout.
> For the case of leaving the bookmarks view, this is a sort of pathological
> situation, since the bookmarks view doesn't really need a document at all.
> But it seems like this could legitimately happen if you are leaving a view that
> happens to have a document where you have not yet reached the end or the body
> element. Maybe there's some simple way to rework the back/forward cache code so
> that we don't attempt this layout when we're no longer associated with a view.

Regarding the last case, see bug 12661 comment #2. Calling clear() earlier did
fix the bug, leaving minor issues with the scrollbars not being reset properly.
However, I suspect that the deeper cause for many, if not all, cases of these
bugs (see also bug 11457) is the regression where documents can enter the page
cache even they haven't finished parsing and loading all their resources. When
you go back to such a document that you left mid-load, it is fetched from the
page cache in its unfinished form and parsing/loading doesn't resume.

I really think at least one of this bug, bug 12661 and bug 11457 should be
closed as a dupe (note that the assertion that's being hit is rather new, with
bug 11457 probably predating it).

Configure bugmail: http://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.

More information about the webkit-unassigned mailing list