manedit crashed with SIGSEGV in gtk_signal_emit()

Description of problem:
I ran manedit with no command line args so I could start building a new man
page. I pressed the "New" button, which triggered an immediate segfault. The
backtrace is as follows, with frames 1-54166 being the same as 54167-54170
repeated over and over. Presumably the segfault is due to running off the end
of the stack.

(gdb) bt -21
#54167 0x00a744d2 in gtk_handlers_run (handlers=0x9a780d8, signal=0xbfa6ed18,
    object=0x9a8ab68, params=0xbfa6ed98, after=1) at gtksignal.c:1917
#54168 0x00a76b92 in gtk_signal_real_emit (object=<value optimized out>,
    signal_id=61, params=0xbfa6ed98) at gtksignal.c:1516
#54169 0x00a76ec4 in gtk_signal_emit_by_name (object=0x9a8ab68,
    name=0x80fdf94 "pressed") at gtksignal.c:618
#54170 0x00a3fdd6 in gtk_marshal_NONE__NONE (object=0x9a8ab68,
    func=0x8064840 <EditorButtonMenuMapCB>, func_data=0x9a8fce8,
    args=0xbfa6f138) at gtkmarshal.c:312
#54171 0x00a744d2 in gtk_handlers_run (handlers=0x9a780d8, signal=0xbfa6f0b8,
    object=0x9a8ab68, params=0xbfa6f138, after=1) at gtksignal.c:1917
#54172 0x00a76b92 in gtk_signal_real_emit (object=<value optimized out>,
    signal_id=61, params=0xbfa6f138) at gtksignal.c:1516
#54173 0x00a77238 in gtk_signal_emit (object=<value optimized out>,
    signal_id=61) at gtksignal.c:552
#54174 0x009cd173 in gtk_button_pressed (button=0x9a8ab68) at gtkbutton.c:318
#54175 0x009ce01d in gtk_button_button_press (widget=0x9a8ab68,
    event=0x9ae2760) at gtkbutton.c:700
#54176 0x00a3fa24 in gtk_marshal_BOOL__POINTER (object=0x9a8ab68,
    func=0x9cdf30 <gtk_button_button_press>, func_data=0x0, args=0xbfa6f508)
    at gtkmarshal.c:28
#54177 0x00a76b16 in gtk_signal_real_emit (object=<value optimized out>,
    signal_id=20, params=0xbfa6f508) at gtksignal.c:1492
#54178 0x00a77238 in gtk_signal_emit (object=<value optimized out>,
    signal_id=20) at gtksignal.c:552
#54179 0x00ab2601 in gtk_widget_event (widget=0x9a8ab68, event=0x9ae2760)
    at gtkwidget.c:2864
#54180 0x00a3db78 in gtk_propagate_event (widget=0x9a8ab68, event=0x9ae2760)
    at gtkmain.c:1378
#54181 0x00a3ecae in gtk_main_do_event (event=0x9ae2760) at gtkmain.c:803
#54182 0x0012975b in gdk_event_dispatch (source_data=0x0,
    current_time=0xbfa6f920, user_data=0x0) at gdkevents.c:2152
#54183 0x0063cd99 in g_main_dispatch (dispatch_time=0xbfa6f920) at gmain.c:656
#54184 0x0063d62d in g_main_iterate (block=<value optimized out>, dispatch=1)
    at gmain.c:877
#54185 0x0063d82f in g_main_run (loop=0x9a2f830) at gmain.c:935
#54186 0x00a3f0ff in gtk_main () at gtkmain.c:524
#54187 0x0808b6dc in main (argc=1, argv=0xbfa6fa94) at main.c:1144

Version-Release number of selected component (if applicable):
manedit-0.8.1-1.fc7.1

How reproducible:
Always

Steps to Reproduce:
1. Start manedit with no command line arguments
2. Press the "New" button

Actual results:
The application momentarily pops up a submenu asking whether to use a template
or not, then crashes with a segfault.

Expected results:
There should be no segfault.

Additional info:

Contacted with upstream....

ManEdit produces this message to stderr just before the segfault:

Gtk-WARNING **: Failed to load module "libgnomebreakpad.so":
libgnomebreakpad.so: cannot open shared object file: No such file or directory

Would you try 0.8.1-2.fc{7,8,9} ?

They can be downloaded from
http://koji.fedoraproject.org/packages/manedit/0.8.1/

Yes, that works great. I'm also getting the warning message identified in
comment #2. Is that anything to worry about? The .so is in
/usr/lib/gtk-2.0/modules.

Well, as currently it seems that manedit is working for now I ignore
the warning. BTW manedit uses GTK+ (version 1.x) so this should have
no relation with /usr/lib/gtk-2.0/modules (version 2.x).

(In reply to comment #4)
> Yes, that works great.

Thank you for confirming!

manedit-0.8.1-2.fc8 has been pushed to the Fedora 8 stable repository. If problems still persist, please make note of it in this bug report.

manedit-0.8.1-2.fc7 has been pushed to the Fedora 7 stable repository. If problems still persist, please make note of it in this bug report.