* gtk/gtktreeview.c (gtk_tree_view_set_destroy_count_func): New
private function for ATK. It notifies you of how many _visible_
children are deleted when a node is deleted. Should be used by
atk only.
Owen Taylor [Tue, 26 Jun 2001 15:15:00 +0000 (15:15 +0000)]
Use new pango_context_get_metrics() to fix problems with font lists in
Tue Jun 26 11:06:34 2001 Owen Taylor <otaylor@redhat.com>
* gtk/gtkclist.c gtk/gtkentry.c gtk/gtkspinbutton.c:
Use new pango_context_get_metrics() to fix problems
with font lists in descriptions. (#56184, reported by
Jonas Borgström)
Tim Janik [Tue, 26 Jun 2001 08:34:26 +0000 (08:34 +0000)]
hm, second attempt. it's fight CVS time again.
Tue Jun 26 09:48:02 2001 Tim Janik <timj@gtk.org>
* Makefile.am (noinst_PROGRAMS): get rid of make-inline-pixbuf
* gdk-pixbuf-data.[hc]: provide gdk_pixbuf_new_from_stream()
instead from gdk_pixbuf_new_from_inline().
* gdk-pixdata.[hc]: auxillary GdkPixdata structure, public
installed API for applications that need to serialize/deserialize
on their own (gimp, BEAST).
* gdk-pixbuf/gdk-pixbuf-csource.c: provide publically installed
program that can dump images in CSource format and Pixbuf
stream format. supports RLE encoding, MACRO formatting etc...
invoke with --help.
Tue Jun 26 10:04:30 2001 Tim Janik <timj@gtk.org>
* gtk/gtkiconfactory.c:
* gtk/gtkitemfactory.c: use gdk_pixbuf_new_from_stream(). fixed up
item factory so inlined pixbufs actually work.
Tim Janik [Tue, 26 Jun 2001 08:32:25 +0000 (08:32 +0000)]
use gdk_pixbuf_new_from_stream(). fixed up item factory so inlined pixbufs
Tue Jun 26 10:04:30 2001 Tim Janik <timj@gtk.org>
* gtk/gtkiconfactory.c:
* gtk/gtkitemfactory.c: use gdk_pixbuf_new_from_stream(). fixed up
item factory so inlined pixbufs actually work.
Tue Jun 26 09:48:02 2001 Tim Janik <timj@gtk.org>
* Makefile.am (noinst_PROGRAMS): get rid of make-inline-pixbuf
* gdk-pixbuf-data.[hc]: provide gdk_pixbuf_new_from_stream()
instead from gdk_pixbuf_new_from_inline().
* gdk-pixdata.[hc]: auxillary GdkPixdata structure, public
installed API for applications that need to serialize/deserialize
on their own (gimp, BEAST).
* gdk-pixbuf/gdk-pixbuf-csource.c: provide publically installed
program that can dump images in CSource format and Pixbuf
stream format. supports RLE encoding, MACRO formatting etc...
invoke with --help.
Added --enable-fbmanager. This is some experimental code that lets several
2001-06-25 Alexander Larsson <alexl@redhat.com>
* configure.in:
Added --enable-fbmanager. This is some experimental code
that lets several GtkFB apps coordinate their access to the
framebuffer.
* acconfig.h:
Added ENABLE_FB_MANAGER.
* gdk/linux-fb/Makefile.am:
Added gdkfbmanager and gdkfbswitch.
* gdk/linux-fb/gdkkeyboard-fb.c:
* gdk/linux-fb/gdkmouse-fb.c:
* gdk/linux-fb/gdkprivate-fb.h:
Split device init and open so that
they can be opened and closed while switched
away.
* gdk/linux-fb/gdkmain-fb.c:
Add the basic manager communication.
* gdk/linux-fb/gdkrender-fb.c:
Don't update to the shadow fb if we're
blocked by the fb manager.
* gtk/gtkdnd.c (gtk_drag_set_icon_default): Use GTK_STOCK_DND
for the default icon. Remove inline XPM.
* gtk/gtkstock.h gtk/gtkiconfactory.c:
Add GTK_STOCK_DND GTK_STOCK_DND_MULTIPLE (Currently, stock_new
is used for GTK_STOCK_DND, but it is a bit too small.)
* gtk/stock-icons/stock_dnd_multiple.png
gtk/stock-icons/Makefile.am: New stock icon moved from gnome-libs.
* gtk/gtkdnd.c (gtk_drag_source_event_cb): Only return
TRUE when starting a drag. In other cases, we are
just observing. (#52995)
* gtk/gtkdnd.[ch] (gtk_drag_set_icon_{stock,pixbuf}): Add
function to set the icon for a drag from a GdkPixbuf
or stock ID.
* gtk/gtkdnd.[ch] (gtk_drag_source_set_icon_{stock,pixbuf}):
Likewise, for drag sources.
* gtk/gtkdnd.h: Deprecate gtk_drag_set_default_icon.
(Now should be done using the stock system.)
Owen Taylor [Sun, 24 Jun 2001 15:34:48 +0000 (15:34 +0000)]
Fix stupid error introduced last night that was making things decidedly
Sun Jun 24 11:29:35 2001 Owen Taylor <otaylor@redhat.com>
* gdk/x11/gdkimage-x11.c (gdk_image_new): Fix stupid error
introduced last night that was making things decidedly not work.
* gtk/*.[ch]: Patch from Kristian Rietveld adding 80 getters
so that we have getter/setter pairing everywhere it makes
sense. (#55767)
* gtk/gtkradiobutton.[ch] gtk/gtktoolbar.c tests/testgtk.:
Rename gtk_radio_button_group to gtk_radio_button_get_group, add a
deprecated compat macro. (#55516)
* gtk/gtklabel.[ch]: Add functions
gtk_label_set/get_use_underline(), gtk_label_set/get_use_markup(),
gtk_label_set_label(), which mirror the property API for GtkLabel.
Make gtk_label_get_attributes() only reflect the attributes
set by gtk_label_set_attributes.
* gtk/gtknotebook.c (gtk_notebook_set_current_page) gtk/gtkcompat.h: Rename
from gtk_notebook_set_page().
Tim Janik [Sat, 23 Jun 2001 20:26:10 +0000 (20:26 +0000)]
disabled g_free (image) calls, as GdkImage is an object. these should be
Sat Jun 23 22:03:20 2001 Tim Janik <timj@gtk.org>
* gdk/x11/gdkimage-x11.c (gdk_image_init): disabled g_free (image)
calls, as GdkImage is an object. these should be g_obejct_unref()
instead but that won't work because of the g_error() statement
in gdk_x11_image_destroy(). so for now, i did s/g_free/LEAK_IMAGE/
to spot these places.
Owen Taylor [Sat, 23 Jun 2001 14:32:44 +0000 (14:32 +0000)]
Fixup some problems with text length handling in error cases.
Sat Jun 23 10:27:53 2001 Owen Taylor <otaylor@redhat.com>
* modules/input/gtkimcontextxim.c: Fixup some problems with
text length handling in error cases.
* gtk/gtkwidget.c (gtk_widget_size_allocate): Reorder
a bit to make test actually warn on attempt to allocation
a negative size. (#56101, Matthias Clasen)
Fix bug noticed by Brian Cameron. (gtk_tree_view_real_expand_row): Fix bug
Fri Jun 22 14:06:08 2001 Jonathan Blandford <jrb@redhat.com>
* gtk/gtktreeview.c (gtk_tree_view_real_collapse_row): Fix bug
noticed by Brian Cameron.
(gtk_tree_view_real_expand_row): Fix bug noticed by Brian
Cameron.
Hans Breuer [Fri, 22 Jun 2001 14:08:51 +0000 (14:08 +0000)]
updated externals
2001-06-22 Hans Breuer <hans@breuer.org>
* gdk/gdk.def : updated externals
* gdk/win32/gdkdrawable-win32.c : corrected some casts
in GDK_NOTEs
* gdk/win32/gdkevents-win32.c : use _gdk_windowing_window_get_offsets
to translate coordinates to the internal > 16 bit system
Try to handle WM_WINDOWPOSCHANGED to get better performance for
when moving/resizing
* gdk/win32/gdkkeys-win32.c : implement gdk_keymap_get_default ()
and gdk_keymap_get_direction (). The latter is untested for
the RTL case
* gtk/gtk.def : updated externals
* gtk/gtkmain.c : gtk_get_default_language () should reurn the result
from pango_language_from_string (), not an already freed pointer
* gtk/stock-icons/stock_menu_sort_ascending.png : PNGs need to be
binary (-kb) to be useable on windoze
Tim Janik [Wed, 20 Jun 2001 04:13:37 +0000 (04:13 +0000)]
make maximum digits compile time configurable via MAX_DIGITS and up to 20.
Wed Jun 20 05:32:05 2001 Tim Janik <timj@gtk.org>
* gtk/gtkspinbutton.c: make maximum digits compile time configurable
via MAX_DIGITS and up to 20. 5 was just ridiculously small.
don't use automatic fixed size buffer for printf-ing floats, doubles
can expand to really _huge_ strings, use g_strdup_printf() instead.
* gtk/gtkarg.[hc], gtk/gtkargcollector.c: got rid of these.
* gtk/gtknotebook.c:
* gtk/gtktable.c:
* gtk/gtkbox.c: ported this over to child properties.
* gtk/gtksettings.c: fetch class properties via
g_object_class_list_properties().
* gtk/gtkcontainer.[hc]: implemented child properties, got rid of the
child arg interface. use gobjectnotifyqueue.c for child property
notification.
* gtk/gtkwidget.[hc]: provide necessary means for container child
properties, i.e. ::child_notify signal,
gtk_widget_freeze_child_notify(),
gtk_widget_child_notify(),
gtk_widget_thaw_child_notify().
* tests/testgtk.c: removed inferior property handling code, for
property editing, a generic module should be used, and GLE
coincidentally fullfills that purpose.
* docs/reference/Makefile.am: disabled gtk docs building, gtk-doc
needs to be adapted to g_object_class_list_properties() before this
builds again.
Rename gtk_tree_sortable_sort_column_id_set_func. It's much shorter now.
Fri Jun 15 18:53:48 2001 Jonathan Blandford <jrb@redhat.com>
* gtk/gtktreesortable.c (gtk_tree_sortable_set_sort_func): Rename
gtk_tree_sortable_sort_column_id_set_func. It's much shorter now.
* gtk/gtktreeview.c (gtk_tree_view_sort_iter_changed): WOOO!!!
Now I can really reorder/sort all Store widgets. treesorttest
seems to just work now.
* gtk/gtklistview.c (gtk_list_store_iter_changed): Whoops.
testtreesort worked through a big coincidence all this time.
* gtk/gtkrbtree.c (_gtk_rbtree_reorder): Nasty bug fix in nasty
code.
* tests/testtreecolumns.c: Big 'ol warning at the top letting
people know that this code should never ever ever be copied.
HA!
While everyone else was busy flaming eash other over configuration
engines, I secretly got sorting completely working with store widgets
and GtkTreeView. GtkTreeModelSort, here I come!!!!!
Whoops. Forgot this a couple months ago. Now GtkTreeStore sort of sorts.
Thu Jun 14 19:21:27 2001 Jonathan Blandford <jrb@redhat.com>
* gtk/gtktreestore.c (gtk_tree_store_sort_helper): Whoops. Forgot
this a couple months ago. Now GtkTreeStore sort of sorts. I'll
finish it off tonight, though it basically works now when it's a
list. Also, I have a sinking feeling neither GtkTreeStore or
GtkListStore actually resort when adding a new item. I'll look
into it.
Owen Taylor [Thu, 14 Jun 2001 21:44:01 +0000 (21:44 +0000)]
Add a GtkWindowGroup struct that allows grouping together multiple windows
Fri Jun 8 17:56:52 2001 Owen Taylor <otaylor@redhat.com>
* gtk/gtkwindow.[ch] gtk/gtkmain.c: Add a GtkWindowGroup struct
that allows grouping together multiple windows so that grabs
within that set of windows only affect those windows.
* gtk/gtkmain.c gtk/gtkwidget.[ch]: Add a "grab_notify"
signal for notification when a widget becomes shadowed
by a grab or is no longer shadowed by a grab.
* gtk/gtkwidget.c (gtk_widget_propagate_state)
gtk/gtkmain.c: (gtk_widget_grab_add): Don't allow
insenstive widgets to maintain a grab.
changed expand_row/collapse_row to be 2 signals each -- test_expand_row
Mon Jun 11 12:43:08 2001 Jonathan Blandford <jrb@redhat.com>
* gtk/gtktreeview.c (gtk_tree_view_class_init): changed
expand_row/collapse_row to be 2 signals each -- test_expand_row
and row_expanded as well as test_collapse_row and row_collapsed.
add read-only "scroll_offset" property for notification when the layout
2001-06-08 Havoc Pennington <hp@redhat.com>
* gtk/gtkentry.c (gtk_entry_class_init): add read-only
"scroll_offset" property for notification when the layout offsets
may have changed
(gtk_entry_adjust_scroll): add notify for scroll_offset
(gtk_entry_layout_index_to_text_index): function to compensate for
preedit string when doing coordinate stuff on the entry's layout
(gtk_entry_text_index_to_layout_index): inverse function
(gtk_entry_get_layout_offsets): hook to get current position of
the layout
(gtk_entry_get_layout): hook to get the layout itself.
* gtk/gtklabel.c (gtk_label_get_layout): new function
* io-gif.c (gif_get_lzw): if delay time is 0 or smaller than
likely timeout resolution, set it to an arbitrarily-chosen short
delay.
Fixes bug #55953 from warmenhoven@yahoo.com
* gtk/gtktextbuffer.c (gtk_text_buffer_remove_all_tags): geez,
don't call g_list funcs on GSList
* gtk/gtkwidget.c (gtk_widget_child_focus): replaces
gtk_container_focus(), since some widgets have focusable locations
which are not other widgets. These widgets should not have to be
containers just to manage the focus. For example, GtkHSV is
currently a container for no good reason. Also, this cleans
up the "if (is_container && is_sensitive && is_drawable)
container_focus ()" mess in widget implementations.
* gtk/gtkcontainer.c (gtk_container_focus): deprecate this
function, and have it just call gtk_widget_child_focus().
* gtk/gtkhsv.c: derive from widget not container
Throughout: fix to reflect GtkContainer::focus change
Owen Taylor [Fri, 8 Jun 2001 16:07:00 +0000 (16:07 +0000)]
Add a direction-changed signal, and gdk_keymap_get_current_direction().
Fri Jun 8 12:03:07 2001 Owen Taylor <otaylor@redhat.com>
* gdk/gdkkeys.[ch]: Add a direction-changed signal,
and gdk_keymap_get_current_direction().
* gdk/x11/gdkevents-x11.c gdk/x11/gdkkeys-x11.c
gdk/x11/gdkmain-x11.c gdk/x11/gdkprivate-x11.h: Track
the current locked group, use it to set the keymap
direction.
* gtk/gtksettings.c: Add a new gtk-split-cursor setting
to determine whether we draw a split cursor or use
a jumping cursor based on the current direction.
* gtk/gtkentry.c gtk/gtktextview.c gtk/gtktextlayout.[ch]:
Obey the split cursor setting.
Fri Jun 8 11:57:50 2001 Owen Taylor <otaylor@redhat.com>
* gtk/gtkimmulticontext.c (activate_cb): Only activate
when toggling on, not when toggling off... (#55906)
Wed Jun 6 10:37:07 2001 Owen Taylor <otaylor@redhat.com>
* gtk/gtkwidget.c (gtk_widget_set_double_buffered):
s/gdk_window_begin_paint/gdk_window_begin_paint_region/
in docs. (#55812, Vitaly Tishkov)
Anders Carlsson [Fri, 8 Jun 2001 07:09:59 +0000 (07:09 +0000)]
Changed signal name from selection_changed to changed in signal connection
2001-06-08 Anders Carlsson <andersca@codefactory.se>
* demos/gtk-demo/main.c (create_tree): Changed signal name from
selection_changed to changed in signal connection to GtkTreeSelection,
also used g_signal_connectc since GtkTreeSelection is now a GObject.
Thu Jun 7 18:25:42 2001 Jonathan Blandford <jrb@redhat.com>
* gtk/gtktreeselection.c: Now it's a GObject instead of a
GtkObject. The GtkTreeSelection::selection_changed signal is now
the GtkTreeSelection::changed signal.
* gtk/gtktreeview.c: Modified to deal with new GtkTreeSelection
object.
* tests/gtktree*.c: Modified to deal with new GtkTreeSelection
object.
fixups for glib-mkenums and glib-genmarshal (not tested yet because with
2001-06-07 Havoc Pennington <hp@redhat.com>
* gtk/Makefile.am: fixups for glib-mkenums and glib-genmarshal
(not tested yet because with absolute path to srcdir I can't build
atk, but it was broken anyway so this may help)
centralize expansion/collapsing so it only happens in one place.
Wed Jun 6 21:18:54 2001 Jonathan Blandford <jrb@webwynk.net>
* gtk/gtktreeview.c (gtk_tree_view_real_collapse_row): centralize
expansion/collapsing so it only happens in one place.
(gtk_tree_view_real_expand_row): ditto.
* configure.in: put G_MODULE_SUPPORTED value into
.pc files
* autogen.sh: support AUTOGEN_SUBDIR_MODE
* Makefile.am: add -uninstalled.pc.in to EXTRA_DIST
2001-06-07 Havoc Pennington <hp@redhat.com>
* pango/Makefile.am: add libpango.la to _DEPENDENCIES for the
other libs
* configure.in: use AM_PATH_GLIB_2_0
* autogen.sh: support AUTOGEN_SUBDIR_MODE
* Makefile.am: dist the .pc.in files
2001-06-07 Havoc Pennington <hp@redhat.com>
* atk/Makefile.am: use @GLIB_GENMARSHAL@ so we can use uninstalled
glib-genmarshal
* configure.in: rearrange the library checks to support
uninstalled linking
* autogen.sh: add support for AUTOGEN_SUBDIR_MODE
* atk-uninstalled.pc.in: new file, allows linking to uninstalled
ATK in giant GTK tarball
2001-06-07 Havoc Pennington <hp@redhat.com>
* tests/Makefile.am: add missing -I flag
* gtk/Makefile.am: use @GLIB_MKENUMS@, @GLIB_GENMARSHAL@, etc.
* configure.in: use pkg-config to see if GModule is
supported; fix to properly turn on included loaders
when GModule isn't supported; don't use AC_CHECK_LIB
when libs are not installed yet
* autogen.sh: add support for AUTOGEN_SUBDIR_MODE
* Makefile.am (SUBDIRS): add m4macros subdir
* gtk/Makefile.am: $(srcdir)/foo targets must be $(srcdir)/foo in
dependencies also.
Wed Jun 6 20:01:38 2001 Jonathan Blandford <jrb@redhat.com>
* gtk/gtktreeview.c
(gtk_tree_view_real_expand_collapse_cursor_row): Handle key
bindings to expand and collapse rows.
(gtk_tree_view_real_select_cursor_parent): New key binding.
(gtk_tree_view_real_toggle_cursor_row): New key binding.
* gtk/gtkmarshal.list: new
marshaller. (VOID__BOOLEAN_BOOLEAN_BOOLEAN)
* gtk/gtkscale.c (gtk_scale_class_init): Change Page Up and Page
Down to move visually rather than logically, since it confuses
people. Add -/+ and Ctrl--/+ bindings for logical movement.
* gtk/gtkspinbutton.c (gtk_spin_button_set_range): clamp the value
to the range that was set
* gtk/gtkrange.c: add value_changed signal, primarily
intended for use with GtkScale
(gtk_range_set_increments): new function
(gtk_range_set_range): new function with weird name
(gtk_range_set_value): new function
(gtk_range_get_value): new function
* gtk/gtkspinbutton.c (gtk_spin_button_get_value): rename
from gtk_spin_button_get_value_as_float(). Compat #define
added for get_value_as_float.
* gtk/gtkhscale.c (gtk_hscale_new_with_range): new function
* gtk/gtkvscale.c (gtk_vscale_new_with_range): new function
2001-06-05 Havoc Pennington <hp@redhat.com>
* test-loaders.c (main): use putenv not setenv, reported by
Armin Theissen
handle case where text_view->layout == NULL by assuming its size is 0,
2001-06-04 Havoc Pennington <hp@redhat.com>
* gtk/gtktextview.c (gtk_text_view_size_request): handle case
where text_view->layout == NULL by assuming its size is 0,
i.e. same as if we haven't done any reflow. Reported by
Hidetoshi Tajima #55448