[Webkit-unassigned] [Bug 150927] Inconsistencies in main resource load delegates when loading from history

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=150927

Michael Catanzaro <mcatanzaro at igalia.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           See Also|                            |https://bugs.webkit.org/sho
                   |                            |w_bug.cgi?id=59111

--- Comment #24 from Michael Catanzaro <mcatanzaro at igalia.com> ---
I just found bug #59111 via the FIXME in WebPageProxy::didCommitLoadForFrame... it is a bit discouraging.

(In reply to comment #23)
> It's still my understanding that the delivery order is wrong, but red
> herring, and that the real issue is that some level of code doesn't
> understand what provisional load vs. committed load means. Sorry, I will not
> have the time to look into this deeper.

To be clear, I was completely wrong; delivery order is not the problem, I only thought it was because I debugged this issue half a year ago and forgot why we had delayed load events. It's my fault for confusing you. The only problem here is that the page cache code sends didCommitLoad too soon, before a main resource is available, like Carlos says.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20151117/65c5663d/attachment.html>