folder widget text entry loves to crash evo

[Jamie Wilkinson]

I'm in evolution mail, and switch folders by clicking on a bolded folder name.
In the previous version, the message window would keep focus, so after loading
the last read mail into the window I could press ] and see the next new message.

The current version, however, keeps the focus on the folder select pane, and
opens up a text entry just below it with a ] in it. Apart from my shock and
confusion when the app didn't behave as expected, by clicking on the message
window to have the text entry disappear and then pressing ], evolution crashes.
 One needs to click *twice* to get the focus to move to the message pane to
prevent the crash.

These folders are on a remote IMAP server.

reproduce like so:
click on folder with new messages
wait for folder to load
press ]
click on message window
press ]
*crash*

http://bugzilla.ximian.com/show_bug.cgi?id=68580: http://bugzilla.ximian.com/show_bug.cgi?id=68580

[Sebastien Bacher (seb128)]

Yes, the box is the new typeahead feature in gtk+.
BTW No crash here, it closes the text entry on the first click and open
it again if I press "["

Could you provide a backtrace of the crash. YOu can get it in the details
of the bug-buddy window or by doing this in a gnome-terminal:
$ gdb evolution
(gdb) run
... use it and get it crashed
(gdb) bt

[Jamie Wilkinson (jaq-spacepants)]

I got a crash here; but it might not be related. I tried the procedure I
specified earlier, run from gdb, but couldn't get it to crash. However, i just
came back to my machine and found it had crashed.

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 16384 (LWP 12532)]
0x0e52e8d4 in g_object_unref () from /usr/lib/libgobject-2.0.so.0
(gdb) bt
#0 0x0e52e8d4 in g_object_unref () from /usr/lib/libgobject-2.0.so.0
#1 0x0e52e8a4 in g_object_unref () from /usr/lib/libgobject-2.0.so.0
#2 0x0e53cfac in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0
#3 0x0e53d3d8 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0
#4 0x0fd394a0 in e_cal_util_remove_instances () from /usr/lib/libecal-1.2.so.0
#5 0x0e53f490 in g_cclosure_marshal_VOID__POINTER ()
   from /usr/lib/libgobject-2.0.so.0
#6 0x0e528820 in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0
#7 0x0e53e150 in g_signal_emit_by_name () from /usr/lib/libgobject-2.0.so.0
#8 0x0e53d0c8 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0
#9 0x0e53d3d8 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0
#10 0x0fd3a2d8 in e_cal_view_start () from /usr/lib/libecal-1.2.so.0
#11 0x0fd141a8 in
_ORBIT_skel_small_GNOME_Evolution_Calendar_CalViewListener_notifyObjectsRemoved
() from /usr/lib/libecal-1.2.so.0
#12 0x0f01279c in ORBit_POA_setup_root () from /usr/lib/libORBit-2.so.0
#13 0x0f017960 in ORBit_OAObject_invoke () from /usr/lib/libORBit-2.so.0
#14 0x0f000544 in ORBit_small_invoke_adaptor () from /usr/lib/libORBit-2.so.0
#15 0x0f012b1c in ORBit_recv_buffer_return_sys_exception ()
   from /usr/lib/libORBit-2.so.0
#16 0x0f012f6c in ORBit_recv_buffer_return_sys_exception ()
   from /usr/lib/libORBit-2.so.0
#17 0x0eff9738 in giop_thread_queue_process () from /usr/lib/libORBit-2.so.0
#18 0x0eff8c64 in giop_thread_same_key () from /usr/lib/libORBit-2.so.0
---Type <return> to continue, or q <return> to quit---
#19 0x0f021a88 in link_servers_move_io_T () from /usr/lib/libORBit-2.so.0
#20 0x0e4512ec in g_main_depth () from /usr/lib/libglib-2.0.so.0
#21 0x0e452800 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0
#22 0x0e452bd0 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0
#23 0x0e4533d0 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#24 0x0f136430 in bonobo_main () from /usr/lib/libbonobo-2.so.0
#25 0x10022804 in main ()
(gdb)

[Sebastien Bacher (seb128)]

(In reply to comment #2)
> I got a crash here; but it might not be related.

This crash seems to be similar to this upstream bug:
http://bugzilla.ximian.com/show_bug.cgi?id=68580

> I tried the procedure I
> specified earlier, run from gdb, but couldn't get it to crash. However, i just
> came back to my machine and found it had crashed.

So you don't get the crash with "[" every time ? How often does it happen ?

[Jamie Wilkinson (jaq-spacepants)]

Well, it was every time, yesterday. Today running inside gdb I couldn't get it
to do it at all, but outside gdb I did manage to crash it in the same way as the
original report.

[Sebastien Bacher (seb128)]

do you still have this bug ?

[Jamie Wilkinson (jaq-spacepants)]

(In reply to comment #5)
> do you still have this bug ?

I havne't used Evo on my laptop (the ubuntu machine this happened on) for two
weeks now, as it was only for work whilst my workstation was being serviced, so
I don't know.

[Jamie Wilkinson (jaq-spacepants)]

(In reply to comment #6)
> (In reply to comment #5)
> > do you still have this bug ?
>
> I havne't used Evo on my laptop (the ubuntu machine this happened on) for two
> weeks now, as it was only for work whilst my workstation was being serviced, so
> I don't know.

I just fired it up, and was able to get it to crash again by hitting ] and
clicking on the message window a few times over, the focus was still on the
folder pane.

Is there a particular version of a library or evo you are referring to?

[Sebastien Bacher (seb128)]

no, but you are the only one to get this bug apparently and I've not real idea
on it for the moment ...

(In reply to comment #7)
> I just fired it up, and was able to get it to crash again by hitting ] and
> clicking on the message window a few times over, the focus was still on the
> folder pane.

how do you click somewhere without moving the focus ? Are you sure that you
don't get this bug ?
http://bugzilla.ubuntu.com/show_bug.cgi?id=3928

[Jamie Wilkinson (jaq-spacepants)]

I'm not seeing this bug anymore, so I reckon it's safe to close.