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 659257 - Gtk-WARNING **: GtkRecentChooserWidget 0x80da890 is mapped but ...
Gtk-WARNING **: GtkRecentChooserWidget 0x80da890 is mapped but ...
Status: RESOLVED FIXED
Product: gtk+
Classification: Platform
Component: Class: GtkRecent
3.1.x
Other Linux
: Normal normal
: ---
Assigned To: gtk-bugs
Emmanuele Bassi (:ebassi)
Depends on:
Blocks:
 
 
Reported: 2011-09-16 15:11 UTC by Kjell Ahlstedt
Modified: 2011-11-16 19:35 UTC
See Also:
GNOME target: ---
GNOME version: ---

Attachments
patch: Maintain map/unmap invariants in GtkRecentChooserDialog (3.07 KB, patch)
2011-11-16 08:16 UTC, Kjell Ahlstedt 		accepted-commit_now Details | Review

Description Kjell Ahlstedt 2011-09-16 15:11:25 UTC 
When a GtkRecentChooserDialog is shown, this warning is printed:
  (testrecentchooser:3094): Gtk-WARNING **: GtkRecentChooserWidget 0x80da890 is
     mapped but visible=1 child_visible=1 parent GtkBox 0x80d9038 mapped=0

   Steps to reproduce:
Start the test program gtk+/tests/testrecentchooser.

   Actual results:
Warning as above. Trace:

+ Trace 228475

  • #0 g_log
    at gmessages.c line 590
  • #1 gtk_widget_verify_invariants
    at gtkwidget.c line 8954
  • #2 gtk_widget_pop_verify_invariants
    at gtkwidget.c line 9015
  • #3 gtk_widget_map
    at gtkwidget.c line 4178
  • #4 gtk_recent_chooser_dialog_map
    at gtkrecentchooserdialog.c line 290
  • #5 g_cclosure_marshal_VOID__VOID
    at gmarshal.c line 85
  • #6 g_type_class_meta_marshal
    at gclosure.c line 885
  • #7 g_closure_invoke
    at gclosure.c line 774
  • #8 signal_emit_unlocked_R
    at gsignal.c line 3202
  • #9 g_signal_emit_valist
    at gsignal.c line 3003
  • #10 g_signal_emit
    at gsignal.c line 3060
  • #11 gtk_widget_map
    at gtkwidget.c line 4173
  • #12 gtk_window_show
    at gtkwindow.c line 4673
  • #13 g_cclosure_marshal_VOID__VOID
    at gmarshal.c line 85
  • #14 g_type_class_meta_marshal
    at gclosure.c line 885
  • #15 g_closure_invoke
    at gclosure.c line 774
  • #16 signal_emit_unlocked_R
    at gsignal.c line 3202
  • #17 g_signal_emit_valist
    at gsignal.c line 3003
  • #18 g_signal_emit
    at gsignal.c line 3060
  • #19 gtk_widget_show
    at gtkwidget.c line 3905
  • #20 gtk_container_show_all
    at gtkcontainer.c line 3175
  • #21 gtk_widget_show_all
    at gtkwidget.c line 4069
  • #22 main
    at testrecentchooser.c line 172 

   Expected results:
No warning.

   Build date and platform: Ubuntu 11.04, source code of gtk+, etc. from git's
      master branch, built with jhbuild on 2011-09-16.

   Additional information:
I suspect that this bug is very similar to bug 640161, fixed with commit
http://git.gnome.org/browse/gtk+/commit/?id=84bf984f8501201e082b888bc5479b5b8770c02b
Comment 1 Murray Cumming 2011-09-27 08:47:40 UTC 
I don't see this warning now, using the latest code from the gtk-3-0 branch. Do you?
Comment 2 Kjell Ahlstedt 2011-09-27 12:17:05 UTC 
I have not tested the gtk-3-0 branch, but I tested with today's version of the
master branch, and the warning remains there. The warning is printed by
gtk_widget_verify_invariants(), and that function is defined within
#ifdef G_ENABLE_DEBUG. The warning is printed only if gtkwidget.c was compiled
with G_ENABLE_DEBUG defined.
Comment 3 Kjell Ahlstedt 2011-11-16 08:16:58 UTC 
Created attachment 201512 [details] [review]
patch: Maintain map/unmap invariants in GtkRecentChooserDialog

This patch deletes gtk_recent_chooser_dialog_map() and
gtk_recent_chooser_dialog_unmap(). It's similar to the corresponding fix in
GtkFileChooserDialog, commit
http://git.gnome.org/browse/gtk+/commit/?id=84bf984f8501201e082b888bc5479b5b8770c02b

With this patch there is no warning from the test program
gtk+/tests/testrecentchooser, and as far I can see there are no bad side
effects.
Comment 4 Emmanuele Bassi (:ebassi) 2011-11-16 12:20:41 UTC 
Review of attachment 201512 [details] [review]:

looks good to me.
Comment 5 Kjell Ahlstedt 2011-11-16 14:59:55 UTC 
Does the result of the patch review mean that I can push my patch to the GTK+
git repository?
Perhaps a silly question, but usually I work only with the C++ bindings (gtkmm,
etc.). I don't know who decide what can be committed to GTK+ and who are
allowed to push commits.
Comment 6 Emmanuele Bassi (:ebassi) 2011-11-16 16:19:33 UTC 
(In reply to comment #5)
> Does the result of the patch review mean that I can push my patch to the GTK+
> git repository?

yes, that's what the accepted-commit_now flag is for. :-)

if you have the commit bit on git.gnome.org, feel free to push the patch to gtk+'s master branch.

> Perhaps a silly question, but usually I work only with the C++ bindings (gtkmm,
> etc.). I don't know who decide what can be committed to GTK+ and who are
> allowed to push commits.

no worries, and thanks for asking.