Re: [orca-list] Update on the situation with closing apps hanging Orca: Please resume testing

[Max <heavy9922 gmail com>]

Hi, Joanye!

Here is another mistake. I do not know whether this error is considered 
unique.
I can reproduce this error so that if need be, give the exact steps for 
reproducing.


13:51:39 - EVENT MANAGER: object:children-changed:remove for [panel | ] 
in [application | gnome-shell] (35, 0, [DEAD])Fatal Python error: 
Segmentation fault

Stack (most recent call first):
  File "/usr/lib/python3.5/site-packages/pyatspi/Accessibility.py", 
line 184 in <lambda>
  File "/usr/local/lib/python3.5/site-packages/orca/event_manager.py", 
line 248 in _queuePrintln
  File "/usr/local/lib/python3.5/site-packages/orca/event_manager.py", 
line 321 in _dequeue
  File "/usr/lib/python3.5/site-packages/pyatspi/registry.py", line 155 
in start
  File "/usr/local/lib/python3.5/site-packages/orca/orca.py", line 575 
in start
  File "/usr/local/lib/python3.5/site-packages/orca/orca.py", line 726 
in main
  File "/usr/local/bin/orca", line 255 in main
  File "/usr/local/bin/orca", line 258 in <module>

Joanmarie Diggs пишет:
> Hey Max.
>
> Thanks again! What would also be helpful is to narrow down the windows
> in question. For instance, if you can always crash it when running
> SeaMonkey, GNOME Terminal, Deluge, Nautilus, and GNOME Shell, quit
> SeaMonkey and try again. If we can come up with a small set of running
> applications in which you can reliably reproduce the crash, then I can
> either add a customization to Orca or send you a patch which tries to
> narrow it down further. Basically I'll try to crash things in a
> different fashion. <grins> Because if we can find a simple and 100%
> reliable means to reproduce this problem, it should make it a bunch
> easier for Mike to fix it for us.
>
> --joanie
>
> On 06/01/2016 06:13 PM, Max wrote:
> > Of course, I will continue testing.
> > At the moment I come across in all cases only mistake I pointed out in
> > my last letter. If I catch something unique then I write about it.
> >
> > Joanmarie Diggs пишет:
> > > Hey Max.
> > >
> > > On 06/01/2016 05:00 PM, Max wrote:
> > >
> > > > Now "orca" really close and does not hang.
> > >
> > > Thanks for the confirmation that it works as expected, and also for the
> > > stack.
> > >
> > > If you look at the top of the stack you provided:
> > >
> > > > "/usr/local/lib/python3.5/site-packages/orca/script_utilities.py", line
> > > > 127 in canBeActiveWindow
> > >
> > > And then look at line 127 in that file, you'll see the call right before
> > > the segfault is "window.clearCache()". That would be this bug:
> > > https://bugzilla.gnome.org/show_bug.cgi?id=767075
> > >
> > > Please keep testing and let me know what you find. So far, I've only
> > > found three AT-SPI2-triggered segfaults. I hope there are not more, but
> > > we might as well try to hunt them all down now.
> > >
> > > --joanie