[Webkit-unassigned] [Bug 236144] [GTK][WPE] Crash at WebKit::bindA11y() in WebKitGTK 2.35.2

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=236144

--- Comment #8 from Carlos Garcia Campos <cgarcia at igalia.com> ---
(In reply to Michael Catanzaro from comment #6)
> (In reply to Carlos Garcia Campos from comment #5)
> > The problem is that the atspi address is no longer an abstract socket and we
> > are assuming it is. I think atspi can use both now, but I'm now sure what it
> > depends on. 
> 
> It should always be a filesystem socket now, since
> https://gitlab.gnome.org/GNOME/at-spi2-core/-/issues/43.

See also https://gitlab.gnome.org/GNOME/at-spi2-core/-/blob/master/bus/at-spi-bus-launcher.c#L501

> > I'm not sure why we try to extract the path from the original
> > socket address but the path is not actually used later for the a11y socket.
> > Note that a11y is also broken in 2.34 with ATK when sandbox is enabled, but
> > there's no crash there. I know how to fix the crash, but not how to make
> > a11y work again with the sandbox enabled.
> 
> It has *never* worked for me, see bug #227640.

It worked for me with both ATK and ATSPI.

-- 
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/20220207/0119e83b/attachment.htm>