Re: [PATCH] g_main_context_pop_thread_default() placement in libsoup

[Dan Winship <danw gnome org>]

On 11/15/2011 06:05 PM, Dmitry Shatrov wrote:
> I realize that there might be a reason for the current placement of
> push/pop calls

Yeah, it used to be necessary to do it that way, but we eventually
decided that this was a bug in GSimpleAsyncResult. That bug was fixed in
glib 2.29.4, and we depend on glib >= 2.30 now, so this patch should be
fine.

Two things:

  1) soup_address_resolve_async() has the same problem, and could
     presumably cause the same leaks in gstreamer if the connection
     was cancelled while the hostname was still being resolved, so
     we need the same fix there.

  2) Could you please submit the patch as a git-format-patch-style
     patch, with a proper commit message and author entry? (eg, see
     https://live.gnome.org/Git/Developers#Contributing_patches).
     Either here or in bugzilla.

Thanks.

-- Dan