[Webkit-unassigned] [Bug 121294] [GTK] socket_embed_hook hitting NULL-check assertion running TestWebKit2

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=121294





--- Comment #5 from Mario Sanchez Prada <mario at webkit.org>  2013-09-20 03:20:41 PST ---
(In reply to comment #4)
> Created an attachment (id=211804)
 --> (https://bugs.webkit.org/attachment.cgi?id=211804&action=review) [details]
> Patch proposal
> 
> After some discussion on #a11y with other GTK+/GAIL developers (Alejandro Pinheiro, Benjamin Otte), we seem to agree the right thing to do is to remove the call to _gtk_accessibility_shutdown() completely, since it made sense in the times of the atk-bridge being loaded as a module, but it does not anymore. Check the full IRC log in [1], which has been finally been marked as duplicate of [2] where a final patch has been proposed and accepted (to commit only after the freeze).
> 
> Hence, I'm proposing now a similar patch here to cope with the issue in WebKitGTK's internal jhbuild in the meanwhile. Please review.
> 
> [1] https://bugzilla.gnome.org/show_bug.cgi?id=708024
> [2] https://bugzilla.gnome.org/show_bug.cgi?id=684076

Ping reviewers?

I think this is quite straightforward fix (even if it's temporary while not in GTK 3.12), and would help get rid of those crazy CRITICALs in the bots.

It will not take much time. Promised :)

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