This patch adds a GtkComboBox:popup-fixed-width to decide if the
popup's width should be a fixed width matching the combo's allocated width.
The patch includes a kindof hack to work around treeviews currently
not supporting height-for-width geometry (for list-mode only), this
hack can be safely removed once treeviews start reporting natural
widths properly.
Add signal GdkWindow::create-surface which allows to use any
surface type as storage for offscreen windows.
Test the new signal in tests/gdkoffscreenbox.c
GtkComboBox now sports a construct-only "has-entry" property which
decides if it uses a GtkEntry to allow additional user input. Also
it has a new "entry-text-column" to fetch strings for the entry
from the model.
This patch deprecates the GtkComboBoxEntry and updates the rest of GTK+
to use the new semantics on GtkComboBox instead.
GtkComboBoxEntry will be removed altogether before GTK+ 3, in a
later commit.
There are two colored boxes with toggle buttons nested
inside several GtkBox. Toggling these to expand mode
should automatically propagate expansion up through
the several GtkBox such that resizing the window
results in resizing the colored boxes.
https://bugzilla.gnome.org/show_bug.cgi?id=628902
If you set a geometry widget via gtk_window_set_geometry_hints() it
becomes very hard to compute appropriate toplevel sizes in pixels
to make the window a particular size. Synthesizing strings and passing
them to gtk_window_parse_geometry() is possible, but to avoid
avoid such ugliness, add functions:
gtk_window_set_default_geometry()
gtk_window_resize_to_geometry()
That act like gtk_window_set_default_size() and
gtk_window_resize() but are in terms of the resize increments of the
geometry widget.
https://bugzilla.gnome.org/show_bug.cgi?id=631796
The geometry widget feature of gtk_window_set_geometry_hints() has
never really worked right because the calculation that GTK+ did to
compute the base size of the window only worked when the geometry
widget had a larger minimum size than anything else in the window.
Setup:
* Move the GtkSizeGroup private functions to a new private header
gtksizegroup-private.h
* Add the possibilty to pass flags to _gtk_size_group_queue_resize(),
with the flag GTK_QUEUE_RESIZE_INVALIDATE_ONLY to suppress adding
the widget's toplevel to the resize queue.
* _gtk_container_resize_invalidate() is added to implement that feature
* _gtk_widget_override_size_request()/_gtk_widget_restore_size_request()
allow temporarily forcing a large minimum size on the geometry
widget without creating resize loops.
GtkWindow:
* Compute the extra width/height around the geometry widget
correctly; print a warning if the computation fails.
* Always make the minimum size at least the natural minimum
size of the toplevel; GTK+ now fails badly with underallocation.
* Always set the base size hint; we were failing to set it
properly when the specified minimum size was overriden, but
it's harmless to always set it.
Tests:
* New test 'testgeometry' that replaces the 'gridded geometry' test
from testgtk. The new test is roughly similar but creates a bunch
of windows showing different possibilities.
* The testgtk test is removed. No need to have both.
https://bugzilla.gnome.org/show_bug.cgi?id=68668
When fitting a widget into its allocation, the second dimension
is always dependent on the first, so gtk_widget_get_preferred_size()
cannot be used directly (because we want the natural height for
the allocated width, not the natural height for the natural width,
which is generally a smaller height than the height-for-minimum-width
or height-for-allocated-width).
Added test to testadjustsize to ensure proper behaviour.
It doesn't make sense to keep them separate as GtkSizeRequest requires a
GtkWidget and GtkWidget implements GtkSizeRequest, so you can never have
one without the other.
It also makes the code a lot easier because no casts are required when
calling functions.
Also, the names would translate to gtk_widget_get_width() and people
agreed that this would be a too generic name, so a "preferred" was added
to the names.
So this patch moves the functions:
gtk_size_request_get_request_mode() => gtk_widget_get_request_mode()
gtk_size_request_get_width() => gtk_widget_get_preferred_width()
gtk_size_request_get_height() => gtk_widget_get_preferred_height()
gtk_size_request_get_size() => gtk_widget_get_preferred_size()
gtk_size_request_get_width_for_height() =>
gtk_widget_get_preferred_width_for_height()
gtk_size_request_get_height_for_width() =>
gtk_widget_get_preferred_height_for_width()
... and moves the corresponding vfuncs to the GtkWidgetClass.
The patch also renames the implementations of the vfuncs in widgets to
include the word "preferrred".
gtk_widget_get_snapshot is on its way out.
The test can be added back once we have gtk_widget_draw(), but until
then, there's no way to take snapshots.
Replace gtk_widget_shape_combine_region() with
gtk_widget_shape_combine_mask() and
gtk_widget_input_shape_combine_region() with
gtk_widget_input_shape_combine_mask().
As GdkBitmap is going away, and the region equivalents already exist,
this seems like pretty much the default step to take.
Includes code to fix up the users.
Dealing with bare pointers is problematic for language bindings,
using interned strings is much more straightforward and more than
good enough for what is needed here.
http://bugzilla.gnome.org/show_bug.cgi?id=630521
Now GtkWrapBox has "horizontal-spreading" and "vertical-spreading" options,
before GtkWrapBox never spread out children across the opposing orientation
(i.e. it never grew "lines" larger then their natural width, they would
act as if set to GTK_WRAP_BOX_SPREAD_START, now they are completely configurable).
When an error occurs with nested traps in place, only the innermost
trap should have the error code stored in it; outer traps are
shielded by the inner trap.
https://bugzilla.gnome.org/show_bug.cgi?id=629608
* don't lose track of previous X error handler
if nested traps are pushed
* free any remaining traps when display
is finalized
Test will fail unless bug 630032 is closed so
gdk_display_close() works.
https://bugzilla.gnome.org/show_bug.cgi?id=630033
This was added to the test only for the sake of making it
easier to reproduce a bug with scrolled windows (bug 629778).
Expected behaviour: The vertical scrollbar should dissapear as soon
as the required height for the full allocation width (without any
vertical scrollbar) is small enough to not need a scrollbar.
Rename the gtk_major_version() etc functions I introduced yesterday to
start with gtk_get. Avoids misleading results in client programs whose
developers that don't notice the change or the compiler warnings, and
when recompiling against gtk3 then use the function addresses as the
version numbers.
The keysyms create a lot of potential namespace conflicts for
C, and are especially problematic for introspection, where we take
constants into the namespace, so GDK_Display conflicts with GdkDisplay.
For C application compatiblity, add gdkkeysyms-compat.h which uses
the old names.
Just one user in GTK+ continues to use gdkkeysyms-compat.h, which is
the gtkimcontextsimple.c, since porting that requires porting more
custom Perl code.
Having variables exported from a DLL is slightly painful and
potentially error-prone on Windows, so let's try get rid of them now
when we can. Starting with these.
Made an enum GtkWrapBoxPacking for the expand/fill horizontal/vertical
boolean options... changed xpadding/ypadding to be horizontal-padding
and vertical-padding for a more consistent api and better readablility.
These have been deprecated and removed from master.
GtkWindow:resizable should be used instead.
This completes commit 1a03a65e36
Reported by Benjamin Otte
Also like the GtkExpander, the label widget is allocated the minimum
height for its allocated width and the remaining space is given to
the child, test case included.
Now the expander requests and distributes space naturally,
the expander widget prioritizes the child widget vertically
and only allocates the minimum height for width to the label
widget.
Includes fixing all callers to use the cairo region API instead. This is
usually just replacing the function names, the only difference is
gdk_region_get_rectangles() being replaced by
cairo_region_num_rectangles() and cairo_region_get_rectangle() which
required a bit more work.
https://bugzilla.gnome.org/show_bug.cgi?id=613284
The old version wasn't introspectable as it didn't have a length
return parameter. Also, delete gtk_tree_path_get_indices_with_depth,
since it's no longer needed.
This commit makes a few massive changes to the extended layout
code:
a.) gtkextendedlayout.c --> gtksizerequest.c
b.) _is_height_for_width --> get_request_mode()
c.) get_desired_size(), get_desired_width(), get_desired_height() -->
get_size(), get_width(), get_height()
This is the first partial commit and only effects portions
of the tree that have already been merged in master (in order to
easily cherry pick this commit).
Conflicts:
gtk/Makefile.am
gtk/gtk.h
gtk/gtk.symbols
gtk/gtkextendedlayout.h
The entire api was deprecated since 2.0. This patch detangles
GtkProgress from GtkProgressBar and moves all the pieces into
GtkProgressBar that are required for non-deprecated progressbar
functionality.
Bug #620618
This is a work in progress to stub out an application class. The
primary goal is to provide a mechanism for applications to export
GtkActions, and there is a standard "Quit" action.
This is based on GApplication.
Future work:
* Add a way to say "This is my application menubar", which gets
put into all toplevel windows on non-OS-X, and into the top
on OS X.
* Support session management.
* Support application settings.
https://bugzilla.gnome.org/show_bug.cgi?id=127958
Use ::keynav-failed for arrow navigation in icon views, so that
it is possible to override error handling. Also add API to get the
row/col of an item. With this, it is possible to make arrow keynav
span adjacent icon views, which is desired in the new control-center
shell. testiconview-keynav demonstrates this.
Bring the various 'run uninstalled' hacks in line with the
new way of doing things, and fix make install for module cache
files.
Patch by Tadej Borovsak.
In particular, rename
- libraries to lib*-3.0.so
- pc files to *-3.0.pc
- include paths to /usr/include/gtk-3.0/*
- module paths to /usr/lib/gtk-3.0/*
- rc files names to gtk-3.0/gtkrc
- commandline utilities to *-3.0
- adjust documentation
Also change the install location for unix-print headers to
/usr/include/gtk-3.0/unix-print/gtk.
This test is a collection of builder files which demonstrate the
added features of natural sizes and height-for-width geometry; so
far it only contains 3 cases, more to come...
Deprecate widget flag macros GTK_WIDGET_STATE, GTK_WIDGET_SAVED_STATE,
GTK_WIDGET_FLAGS, GTK_WIDGET_TOPLEVEL, GTK_WIDGET_NO_WINDOW and
GTK_WIDGET_COMPOSITE_CHILD.
Also deprecate the type macros GTK_WIDGET_TYPE, GTK_OBJECT_TYPE_NAME and
GTK_OBJECT_TYPE which have become redundant.
Instances of GTK_WIDGET_TOPLEVEL are replaced with gtk_widget_is_toplevel,
GTK_WIDGET_TYPE is replaced with G_OBJECT_TYPE, GTK_WIDGET_COMPOSITE_CHILD
is replaced with use of the "composite-child" property and uses of
GTK_WIDGET_NO_WINDOW are adjusted to use gtk_widget_get_has_window.
Uses of GTK_WIDGET_SAVED_STATE and GTK_WIDGET_FLAGS inside GtkWidget are
changed to direct flag usage.
Documentation is updated to refer to gtk_widget_set_has_window and
gtk_widget_get_has_window.
Gail and tests are updated as well.
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=69872
This adds LDFLAGS everywhere where they were previously pulled in via
other libraries. This is however unsupported by modern linkers.
You can trigger these failures by building with gold (or, I'm told, with
very new ld).
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=604289
In order to respect this properly, we have to restore the original
behavior by watching a flag to check if the user has already set this.
* gtk/gtkassistant.c: fix the bug by introducing a flag to check for
non-automatic value setting
* tests/testassistant.c: updated the "generous assistant" to permit
quick manual testing of this feature
Add always-show-image="true|false" attribute to <menuitem> element in UI
manager markup. If present, it overrides the always-show-image property
on image menu items. Bug #589842.
Either g_type_register_static_simple (used by G_DEFINE_TYPE_EXTENDED)
and G_IMPLEMENT_INTERFACE use automatic variables for GTypeInfo and
GInterfaceInfo structs, while tutorials and source code often use
static variables. This commit consistently adopts the former method.
https://bugzilla.gnome.org/show_bug.cgi?id=600158
we now use gdk_offscreen_window_set_embedder() instead of a signal
to get the parent. This also replaces set_has_offscreen_changes.
Rename "parent" in all embedding related names to "embedder" to make it
more obviously different than the normal parent.
Rename gdk_window_get_offscreen_pixmap to gdk_offscreen_window_get_pixmap
to match the other offscreen calls.
Rename gdk_window_offscreen_children_changed to gdk_window_geometry_changed
as this is more descriptive.
We can use this to test bugs that happen when setting a filename before the
file chooser is shown, as in bgo#161670
Signed-off-by: Federico Mena Quintero <federico@novell.com>
This also removes the (unimplemented) possibility to change a window to
non-native. This seems generally not very useful, and there are some problems
with it, for instance if two "users" need a window to be native and then
one of the "users" doesn't need it anymore it can't change it back, because
it is unaware of the other reason the window is native.
2009-02-19 Michael Natterer <mitch@imendio.com>
* tests/testgtk.c: add some orientation flipping code to the
"flipping" test.
svn path=/trunk/; revision=22376
Bug 565656 – Add marks to scales
* gtk/gtkrange.[hc]: Add internal api to define 'stop values'
that have a little resistance when dragging the slider over it.
* gtk/gtk.symbols:
* gtk/gtkscale.[hc] (gtk_scale_add_mark): New function to add
a 'mark' to a scale, which will draws a tick, plus optionally
some text, and makes the value a stop value.
(gtk_scale_clear_values): Removes all marks.
* tests/testscale.c: Test for marks on scales
* tests/Makefile.am: Integrate it
svn path=/trunk/; revision=22149
* gtk/gtktoolbutton.c: Properly create a menu proxy from a GIcon.
Patch by Christian Persch
* tests/testtoolbar.c: Add an example with a GIcon
svn path=/trunk/; revision=22084
2008-12-26 Matthias Clasen <mclasen@redhat.com>
* gtk/gtkentry.c: Don't emit ::icon-pressed on nonactivatable
icons. Fix up docs to match actual api.
* tests/testentryicons.c: Reshuffle tests a bit. Add a DND test.
svn path=/trunk/; revision=21942
2008-12-19 Cody Russell <bratsche@gnome.org>
Bug 85292 – add an icon to gtkentry
* gtk/gtkmarshalers.list: Add VOID:INT,BOXED
* tests/testentryicons.c: Initial icon entry test
* tests/Makefile.am: Add testentryicons
* gtk/gtkentry.[ch]: Add API for setting primary/secondary icons
and other features related to them.
svn path=/trunk/; revision=21914
2008-12-05 Michael Natterer <mitch@imendio.com>
Bug 546285 – Allow GtkEntry to draw progress
* gtk/gtkentry.[ch]: add new API similar to GtkProgressBar which
allows to set the entry's progress_fraction, its progress_pulse_step
and to let the entry's progress pulse.
* gtk/gtk.symbols: updated.
* tests/testgtk.c: add progress demo code to the "Entry" window.
svn path=/trunk/; revision=21846
2008-11-26 Christian Dywan <christian@imendio.com>
Bug 561504 – testgtk should load rc file from sub folder
* tests/testgtk.c (main):
Make testgtk look in subfolder and warn if not found
svn path=/trunk/; revision=21809
2008-11-01 Tor Lillqvist <tml@novell.com>
* tests/testclientmessage.c: New interactive test program to
verify client message functionality.
* tests/Makefile.am: Add it.
svn path=/trunk/; revision=21749
2008-10-13 Cody Russell <cody@jhu.edu>
* test/testfilechooser.c: Fix option parsing so that -a and
--action work correctly.
svn path=/trunk/; revision=21645
2008-08-05 Michael Natterer <mitch@imendio.com>
* tests/testiconview.c: don't use the deprecated
gtk_box_pack_start_defaults().
svn path=/trunk/; revision=21010
2008-08-04 Michael Natterer <mitch@imendio.com>
* tests/prop-editor.c: undeprecate (apart from still using
GtkOptionMenu).
svn path=/trunk/; revision=20979
2008-08-01 Hans Breuer <hans@breuer.org>
* gdk/makefile.msc : build gdkwindowimpl
* gtk/gtkwidget.c : fix c99ism - declaration needs to be at the
start of a block
* tests/makefile.msc : update
svn path=/trunk/; revision=20926
2008-07-21 Michael Natterer <mitch@imendio.com>
Bug 442042 – GtkScaleButton is too limited
* tests/testvolumebutton.c: forgot to commit code that tests
toggling the button's orientation. Also fixes the file's coding
style.
svn path=/trunk/; revision=20882
2008-07-11 Kristian Rietveld <kris@gtk.org>
Bug 316087 - Resizing columns is chaotic
* gtk/gtktreeprivate.h: add new member fields.
* gtk/gtktreeview.c (gtk_tree_view_init), (validate_row): set post
validation flag,
(gtk_tree_view_size_allocate_columns): rework the size allocation
mechanism to only recalculate the expand values if the width of the
widget, content or the column configuration has changed,
(gtk_tree_view_size_allocate): move call to size_allocate_columns()
to before the adjustment updates so the proper width is used after
we updated it,
(gtk_tree_view_button_press), (gtk_tree_view_motion_resize_column):
use the column width minus the expand value for the resized width,
(gtk_tree_view_move_column_after): update call to
gtk_tree_view_size_allocate_columns().
* gtk/gtktreeviewcolumn.c (gtk_tree_view_column_set_expand): set use
resized width to FALSE.
* tests/Makefile.am:
* tests/testtreecolumnsizing.c: new interactive test program
for testing column resizing with different column configurations.
svn path=/trunk/; revision=20818
2008-06-30 Cody Russell <bratsche@gnome.org>
* Practically everything changed.
Change all references of GIMP Toolkit (and variations of it)
to GTK+ Toolkit, showing no mercy at all to our beloved
ancestry. (#540529)
svn path=/trunk/; revision=20709
2008-06-20 Hans Breuer <hans@breuer.org>
* gtk/gtkmenuitem.c gtk/gtkmessagedialog.c gtk/gtkplug.c : must return
a value of the appropriate type with g_return_val_if_fail
* gdk/gdkconfig.h.win32 : add the GSEAL definition here as well
* gtk/gtkfilesystem.c : don't add the "File System" on win32
* tests/makefile.msc : remove broken autotestfile(system|chooser)
from build
svn path=/trunk/; revision=20657
2008-06-20 Mikael Hallendal <micke@imendio.com>
* tests/testthreads.c: Use g_signal_connect instead of
gtk_signal_connect.
* demos/testgtk/main.c: switched to use GSignal instead of GtkSignal.
While the code is not built, people may still look at it.
svn path=/trunk/; revision=20654
2008-06-18 Michael Natterer <mitch@imendio.com>
* gtk/gtktypeutils.[ch]: deprecate gtk_type_class().
* gtk/gtk.symbols: changed accordingly.
* gtk/gtksettings.c (gtk_settings_install_property)
(gtk_settings_install_property_parser): get rid of using
gtk_type_class() by keeping a static reference to GtkSettingsClass
around in both functions and passing that pointer on to
settings_install_property_parser().
* tests/testgtk.c: use g_type_class_ref()/unref() instead of
gtk_type_class().
* gtk/gtkobject.c: #undef GTK_DISABLE_DEPRECATED around including
gtktypeutils.h.
svn path=/trunk/; revision=20444
2008-05-27 Matthias Clasen <mclasen@redhat.com>
Bug 522084 – GIcon support for GtkIconTheme & GtkImage
* gtk/gtkicontheme.[hc]: Add support for GIcon based lookups.
* gtk/gtkimage.[hc]: Allow setting an image from a GIcon.
* gtk/gtk.symbols: Additions
* tests/testimage.c: Add a GIcon test
svn path=/trunk/; revision=20236
2008-05-29 Michael Natterer <mitch@imendio.org>
* gtk/Makefile.am
* gtk/gtkunixprint.h: add a single-include header for the unix
print stuff.
* gtk/gtkpagesetupunixdialog.h
* gtk/gtkprinter.h
* gtk/gtkprintjob.h
* gtk/gtkprintunixdialog.h: add single-include guards that #error
out if GTK_DISABLE_SINGLE_INCLUDES is defined and any of these
files is included individually.
* gtk/gtkprintbackend.h
* gtk/gtkprinter-private.h
* modules/printbackends/cups/gtkprintbackendcups.c
* modules/printbackends/cups/gtkprintercups.h
* modules/printbackends/lpr/gtkprintbackendlpr.c
* tests/buildertest.c
* tests/defaultvaluetest.c
* docs/tools/widgets.c: include <gtk/gtkunixprint.h> instead of
individual files.
svn path=/trunk/; revision=20230
2008-05-25 Jan Arne Petersen <jpetersen@jpetersen.org>
* gtk/gtkliststore.c: (list_store_text): Use type of the correct
column (#529535).
* tests/buildertest.c: (test_list_store): Add test.
svn path=/trunk/; revision=20142
2008-03-18 10:49:20 Tim Janik <timj@imendio.com>
* Applied pixmap redirection patch by Alexander Larsson with
various updates from:
Bug 318807 – Offscreen windows and window redirection.
Updates:
* updated docs to mention "Since 2.16".
* tests/testgtk.c: fixed snapshooting pixmap leak.
convert pixmap to pixbuf after snapshooting, to compensate for different
bit depths (occurs when snapshooting ARGB visuals and displaying the
pixmap in an RGB visual).
* gdk/gdkwindow.[hc]: made GdkWindowRedirect private.
* gdk/gdkwindow.c: removed damage idle handler, there's no aparent
need for it. enqueue damage notification as GDK_DAMAGE events
for each painting redirection at the start of the event queue.
consider windows with a redirection fully visible when invalidating,
and when updating from backing store. cleaned up stale variables.
* gdk/gdkevents.c: added _gdk_event_queue_prepend().
* gtk/gtkwidget.c: fixed coordinates for !NO_WINDOW widgets in
gtk_widget_get_snapshot; this fixes garbage snap offsets for gammacurve,
tree, drawingarea, text, handlebox, etc.
clip the redirected window hierarchy to window sizes, the visible
rectangles don't need to be taken into account here.
extended snapshooting docs to recommend gdk_pixbuf_get_from_drawable()
in case pixmap visuals could mismatch.
* gdk/x11/gdkwindow-x11.c: removed _gdk_windowing_window_get_visible_rect().
Base patch:
* tests/testgtk.c: add a "Snapshot" test to demonstrate snapshooting
of possibly obscured widgets into an offscreen pixmap.
* gtk/gtkwidget.[hc]: add GtkWidget::damage-event signal, add
gtk_widget_get_snapshot() to render a widget's contents to a GdkPixmap.
* gtk/gtkmain.c: dispatch GDK_DAMAGE events.
* gdk/gdkwindow.c: moved outer gdk_window_new() and gdk_window_reparent()
implementations here, adapted them to propagate redirects to child windows.
gdk_window_end_paint(): copy repainted window contents to redirection pixmap,
clipped to visible region. queue GDK_DAMAGE event delivery.
gdk_window_redirect_to_drawable(): install window painting redirection.
gdk_window_remove_redirection(): remove previously installed redirection.
* gdk/x11/gdkwindow-x11.c: added _gdk_windowing_window_get_visible_rect(),
renamed _gdk_window_new() and _gdk_window_reparent().
* gdk/gdkwindow.h: added GdkWindowRedirect* to GdkWindowObject, export
gdk_window_redirect_to_drawable() and gdk_window_remove_redirection().
* gdk/gdkevents.h: added GDK_DAMAGE event type.
* gdk/gdkevents.c: extract time and state from GDK_DAMAGE events.
* gdk/gdkinternals.h: added internal prototypes.
svn path=/trunk/; revision=20122
2008-04-02 Björn Lindqvist <bjourne@gmail.com>
* tests/autotestfilechooser.c: Cover
GTK_FILE_CHOOSER_ACTION_CREATE_FOLDER action in tests for current
name. (#346058, Joe Halliwell)
svn path=/trunk/; revision=19967
2008-03-20 Tor Lillqvist <tml@novell.com>
* tests/testfilechooserbutton.c (chooser_update_preview_cb): Don't
pass NULL for %s to g_message().
(main): Drop unnecessary call to gtk_widget_destroy().
svn path=/trunk/; revision=19908
2008-03-17 Christian Kellner <gicmo@gnome.org>
Implement GtkMountOperation, a subclass of GMountOperation
to be used with gio wherever there is the need to ask the
user for credentials or questions while mounting a volume.
This is bug #522245
* gtk/gtkmountoperation.c:
* gtk/gtkmountoperation.h:
Implement GtkMountOperation.
* gtk/gtk.h: Add gtkmountoperation.h
* gtk/Makefile.am: Add gtkmountoperation.[hc]
* gtk/gtk.symbols: Add symbols of GtkMountOperation.
* tests/testmountoperation.c: Test program for it.
* tests/Makefile.am: Add testmountoperation.
svn path=/trunk/; revision=19894
2008-03-10 Tor Lillqvist <tml@novell.com>
* tests/teststatusicon.c (do_exit, popup_menu): Add the
possibility to test what happens if the process just calls exit()
without any orderly cleanup of GTK+.
svn path=/trunk/; revision=19745
