[Webkit-unassigned] [Bug 194131] Load events may occur in wrong order when JS redirects page before subresource load finishes

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=194131

Michael Catanzaro <mcatanzaro at igalia.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |achristensen at apple.com,
                   |                            |mcatanzaro at igalia.com,
                   |                            |rniwa at webkit.org

--- Comment #1 from Michael Catanzaro <mcatanzaro at igalia.com> ---
After a few hours staring at FrameLoader.cpp, I think the current behavior of WebCore is OK -- because the load really never finishes -- and we could choose to just run the finished event at the WPE/GTK API level. Alternatively, it looks like FrameLoader::setState would be a good place to run m_client.dispatchDidFinishLoad to fake the finish, but I'd worry about unintended side effects. CCing people who've touched loader and NavigationClient for opinions on intended behavior.

P.S. In this example, didFinishDocumentLoadForFrame is executed, but the subresource never finishes because there's no HTTP server running on www.example.com:8080, so didFinishLoadForFrame never runs. (We have to use a valid host to ensure it resolves, or the load will fail.)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20190201/6ceda2c7/attachment-0001.html>