Bug#758619: more complete backtrace

[Simon McVittie]

On 21/09/14 22:47, Sandro Tosi wrote:
> On Thu, Aug 21, 2014 at 10:38 AM, Simon McVittie <smcv at debian.org> wrote:
>> +gtk.set_interactive (0)
>>  gtk.gdk.threads_init ()
> 
> sadly this patch didnt fix the problem, and I can replicate it in a
> clean sid chroot. do you have any other suggestions?

The next best ideas I have are:

* ask the gtk2 maintainers to apply upstream commit
  fbf38d16bcc26630f0f721d266509f5bc292f606 (attached) to work around
  pygtk2 being an example of the "wrong code" mentioned in that commit;
  and/or

* ask the pygtk2 maintainers to stub out the "interactive" code in
  pygtk2, slightly breaking interactive use of pygtk2 but maybe nobody
  will actually notice; and/or

* switch reportbug to using Gtk3 via python-gi

I realise that last one is not trivial, but pygtk2 is dead upstream and
gtk2 is not a whole lot better, so it would be a good idea for jessie+1
regardless.

    S

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-threads-Do-not-release-the-GDK-lock-if-it-hasn-t-bee.patch
Type: text/x-patch
Size: 2495 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/pkg-gnome-maintainers/attachments/20140922/95b366df/attachment-0001.bin>