Bug 325308 – gThumb crash when reaching the end of a directory during fullscreen viewing

After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.

Bug 325308 - gThumb crash when reaching the end of a directory during fullscreen viewing


Summary:	gThumb crash when reaching the end of a directory during fullscreen viewing


Status:	RESOLVED FIXED

Product:	gthumb
Classification:	Other
Component:	general
Version:	2.7.x
Hardware:	Other other

Importance:	High critical
Target Milestone:	---
Assigned To:	Paolo Bacchilega
QA Contact:	Paolo Bacchilega

URL:
Whiteboard:

Duplicates:	328995 335336 (view as bug list)
Depends on:
Blocks:

Reported:	2005-12-30 16:21 UTC by Benjamin Dauvergne
Modified:	2006-03-22 08:18 UTC

See Also:
GNOME target:	---
GNOME version:	2.13/2.14

Description Benjamin Dauvergne 2005-12-30 16:21:23 UTC

Distribution: Ubuntu 6.04 (dapper)
Package: gthumb
Severity: Normal
Version: GNOME2.13.3 2.7.1
Gnome-Distributor: Ubuntu
Synopsis: gThumb crash when reaching the end of a directory during fullscreen viewing
Bugzilla-Product: gthumb
Bugzilla-Component: general
Bugzilla-Version: 2.7.1
BugBuddy-GnomeVersion: 2.0 (2.13.0)
Description:
How often does this happen?
Always.



Debugging Information:

Backtrace was generated from '/usr/bin/gthumb'

(no debugging symbols found)
Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1".
(no debugging symbols found)
(no debugging symbols found)
(no debugging symbols found)
(no debugging symbols found)
(no debugging symbols found)
[Thread debugging using libthread_db enabled]
[New Thread -1209067840 (LWP 5645)]
[New Thread -1299194960 (LWP 5662)]
[New Thread -1290802256 (LWP 5661)]
[New Thread -1282409552 (LWP 5660)]
[New Thread -1263178832 (LWP 5659)]
[New Thread -1272411216 (LWP 5658)]
[New Thread -1253934160 (LWP 5652)]
[New Thread -1245541456 (LWP 5651)]
[New Thread -1237148752 (LWP 5650)]
[New Thread -1228756048 (LWP 5649)]
[New Thread -1228461136 (LWP 5648)]
[New Thread -1219822672 (LWP 5647)]
[New Thread -1211409488 (LWP 5646)]
0xffffe410 in __kernel_vsyscall ()

+ Trace 64864

Thread 1 (Thread -1209067840 (LWP 5645))

#0 __kernel_vsyscall
#1 __waitpid_nocancel
from /lib/tls/i686/cmov/libpthread.so.0
#2 libgnomeui_module_info_get
from /usr/lib/libgnomeui-2.so.0
#3 <signal handler called>
#4 gdk_screen_get_rgb_colormap
from /usr/lib/libgdk-x11-2.0.so.0
#5 gdk_draw_rgb_image_dithalign
from /usr/lib/libgdk-x11-2.0.so.0
#6 image_viewer_is_frame_visible
from /usr/lib/libgthumb.so
#7 image_viewer_is_frame_visible
from /usr/lib/libgthumb.so
#8 image_viewer_is_frame_visible
from /usr/lib/libgthumb.so
#9 image_viewer_scroll_to
from /usr/lib/libgthumb.so
#10 image_viewer_scroll_to
from /usr/lib/libgthumb.so
#11 gthumb_marshal_VOID__ENUM_ENUM
from /usr/lib/libgthumb.so
#12 g_cclosure_new_swap
from /usr/lib/libgobject-2.0.so.0
#13 g_closure_invoke
from /usr/lib/libgobject-2.0.so.0
#14 g_signal_stop_emission
from /usr/lib/libgobject-2.0.so.0
#15 g_signal_emitv
from /usr/lib/libgobject-2.0.so.0
#16 gtk_bin_get_child
from /usr/lib/libgtk-x11-2.0.so.0
#17 gtk_binding_set_by_class
from /usr/lib/libgtk-x11-2.0.so.0
#18 gtk_binding_set_by_class
from /usr/lib/libgtk-x11-2.0.so.0
#19 gtk_bindings_activate
from /usr/lib/libgtk-x11-2.0.so.0
#20 image_loader_load_from_image_loader
from /usr/lib/libgthumb.so
#21 _gtk_marshal_BOOLEAN__BOXED
from /usr/lib/libgtk-x11-2.0.so.0
#22 g_cclosure_new_swap
from /usr/lib/libgobject-2.0.so.0
#23 g_closure_invoke
from /usr/lib/libgobject-2.0.so.0
#24 g_signal_stop_emission
from /usr/lib/libgobject-2.0.so.0
#25 g_signal_emit_valist
from /usr/lib/libgobject-2.0.so.0
#26 g_signal_emit
from /usr/lib/libgobject-2.0.so.0
#27 gtk_widget_activate
from /usr/lib/libgtk-x11-2.0.so.0
#28 gtk_window_propagate_key_event
from /usr/lib/libgtk-x11-2.0.so.0
#29 gtk_window_activate_key
from /usr/lib/libgtk-x11-2.0.so.0
#30 _gtk_marshal_BOOLEAN__BOXED
from /usr/lib/libgtk-x11-2.0.so.0
#31 g_cclosure_new_swap
from /usr/lib/libgobject-2.0.so.0
#32 g_closure_invoke
from /usr/lib/libgobject-2.0.so.0
#33 g_signal_stop_emission
from /usr/lib/libgobject-2.0.so.0
#34 g_signal_emit_valist
from /usr/lib/libgobject-2.0.so.0
#35 g_signal_emit
from /usr/lib/libgobject-2.0.so.0
#36 gtk_widget_activate
from /usr/lib/libgtk-x11-2.0.so.0
#37 gtk_propagate_event
from /usr/lib/libgtk-x11-2.0.so.0
#38 gtk_main_do_event
from /usr/lib/libgtk-x11-2.0.so.0
#39 _gdk_events_queue
from /usr/lib/libgdk-x11-2.0.so.0
#40 g_main_context_dispatch
from /usr/lib/libglib-2.0.so.0
#41 g_main_context_check
from /usr/lib/libglib-2.0.so.0
#42 g_main_loop_run
from /usr/lib/libglib-2.0.so.0
#43 gtk_main
from /usr/lib/libgtk-x11-2.0.so.0
#44 main
#0 __kernel_vsyscall





------- Bug created by bug-buddy at 2005-12-30 16:21 -------


Unknown version 2.7.1 in product gthumb.  Setting version to "2.7.x".

Comment 1 Benjamin Dauvergne 2005-12-30 23:29:51 UTC

It seems to me that load_prev_image and load_next_image in gth-fullscreen.c can trigger gtk_window_close on the fullscreen window.

In viewer_key_press_cb, there is:

||         case GDK_Escape:
||         case GDK_q:
||         case GDK_v:
||         case GDK_f:
||         case GDK_F11:
--                 gth_window_close (window);
34                 retval = TRUE;
34                 break;

retval is set to TRUE, so the event is not propagated and nothing more can happer to window.

but later there is:
34                 /* Load next image. */
||         case GDK_space:
||         case GDK_n:
||         case GDK_Page_Down:
--                 load_next_image (fullscreen);
34                 retval = TRUE;
34                 break;
34
34                 /* Load previous image. */
||         case GDK_p:
||         case GDK_b:
||         case GDK_BackSpace:
||         case GDK_Page_Up:
--                 load_prev_image (fullscreen);
34                 break;
34
34                 /* Show first image. */

retval is not changed so that it finished to be treated by 
image_viewer_key_down that ask for a scroll of a widget whose window has been destroyed.

if i add "retvall = TRUE;" after "load_next_image" the problem disappear.

Comment 2 Paolo Bacchilega 2006-01-04 22:27:30 UTC

I commited a patch that blocks event propagation by default as you suggested, and removed Left and Right as shortcuts to move to the next/prev image as they are already used to scroll the image horizontally.

thanks for the bug report.

Comment 3 Christian Kirbach 2006-01-28 18:08:30 UTC

*** Bug 328995 has been marked as a duplicate of this bug. ***

Comment 4 Paolo Bacchilega 2006-03-22 08:18:59 UTC

*** Bug 335336 has been marked as a duplicate of this bug. ***