gtk2/gtk/gtksignal.c

2010 lines
49 KiB
C
Raw Normal View History

1997-11-24 22:37:52 +00:00
/* GTK - The GIMP Toolkit
* Copyright (C) 1995-1997 Peter Mattis, Spencer Kimball and Josh MacDonald
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
1997-11-24 22:37:52 +00:00
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
1997-11-24 22:37:52 +00:00
*/
/*
* Modified by the GTK+ Team and others 1997-1999. See the AUTHORS
* file for a list of people on the GTK+ Team. See the ChangeLog
* files for a list of changes. These files are distributed with
* GTK+ at ftp://ftp.gtk.org/pub/gtk/.
*/
1997-11-24 22:37:52 +00:00
#include <stdarg.h>
#include <string.h>
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
#include <stdio.h>
1997-11-24 22:37:52 +00:00
#include "gtksignal.h"
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
#include "gtkargcollector.c"
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
#define SIGNAL_BLOCK_SIZE (100)
#define HANDLER_BLOCK_SIZE (200)
#define EMISSION_BLOCK_SIZE (100)
#define DISCONNECT_INFO_BLOCK_SIZE (64)
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
#define MAX_SIGNAL_PARAMS (31)
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
enum
{
EMISSION_CONTINUE,
EMISSION_RESTART,
EMISSION_DONE
};
1997-11-24 22:37:52 +00:00
#define GTK_RUN_TYPE(x) ((x) & GTK_RUN_BOTH)
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
typedef struct _GtkSignal GtkSignal;
typedef struct _GtkSignalHash GtkSignalHash;
typedef struct _GtkHandler GtkHandler;
typedef struct _GtkEmission GtkEmission;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
typedef struct _GtkEmissionHookData GtkEmissionHookData;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
typedef struct _GtkDisconnectInfo GtkDisconnectInfo;
1997-11-24 22:37:52 +00:00
typedef void (*GtkSignalMarshaller0) (GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer data);
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
struct _GtkSignal
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
GtkType object_type;
gchar *name;
guint function_offset;
GtkSignalMarshaller marshaller;
GtkType return_val;
guint signal_flags : 16;
guint nparams : 16;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkType *params;
GHookList *hook_list;
1997-11-24 22:37:52 +00:00
};
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
struct _GtkSignalHash
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkType object_type;
GQuark quark;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
1997-11-24 22:37:52 +00:00
};
struct _GtkHandler
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint id;
GtkHandler *next;
GtkHandler *prev;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint blocked : 20;
guint object_signal : 1;
guint after : 1;
guint no_marshal : 1;
guint16 ref_count;
guint16 signal_id;
GtkSignalFunc func;
gpointer func_data;
GtkSignalDestroy destroy_func;
};
1997-11-24 22:37:52 +00:00
struct _GtkEmission
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkObject *object;
guint16 signal_id;
guint in_hook : 1;
GtkEmission *next;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
};
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
struct _GtkEmissionHookData
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkObject *object;
guint signal_id;
guint n_params;
GtkArg *params;
};
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
struct _GtkDisconnectInfo
{
GtkObject *object1;
guint disconnect_handler1;
guint signal_handler;
GtkObject *object2;
guint disconnect_handler2;
1997-11-24 22:37:52 +00:00
};
static guint gtk_signal_hash (gconstpointer h);
static gint gtk_signal_compare (gconstpointer h1,
gconstpointer h2);
1997-11-24 22:37:52 +00:00
static GtkHandler* gtk_signal_handler_new (void);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static void gtk_signal_handler_ref (GtkHandler *handler);
static void gtk_signal_handler_unref (GtkHandler *handler,
GtkObject *object);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static void gtk_signal_handler_insert (GtkObject *object,
1997-11-24 22:37:52 +00:00
GtkHandler *handler);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static void gtk_signal_real_emit (GtkObject *object,
guint signal_id,
GtkArg *params);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static guint gtk_signal_connect_by_type (GtkObject *object,
guint signal_id,
1997-11-24 22:37:52 +00:00
GtkSignalFunc func,
gpointer func_data,
GtkSignalDestroy destroy_func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gint object_signal,
gint after,
gint no_marshal);
static guint gtk_alive_disconnecter (GtkDisconnectInfo *info);
static GtkEmission* gtk_emission_new (void);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
static void gtk_emission_add (GtkEmission **emissions,
1997-11-24 22:37:52 +00:00
GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_type);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
static void gtk_emission_remove (GtkEmission **emissions,
1997-11-24 22:37:52 +00:00
GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_type);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
static gint gtk_emission_check (GtkEmission *emissions,
1997-11-24 22:37:52 +00:00
GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_type);
static gint gtk_handlers_run (GtkHandler *handlers,
GtkSignal *signal,
GtkObject *object,
GtkArg *params,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gint after);
static gboolean gtk_emission_hook_marshaller (GHook *hook,
gpointer data);
static gboolean gtk_signal_collect_params (GtkArg *params,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint nparams,
GtkType *param_types,
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
GtkType return_type,
va_list var_args);
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
#define LOOKUP_SIGNAL_ID(signal_id) ( \
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
signal_id > 0 && signal_id < _gtk_private_n_signals ? \
(GtkSignal*) _gtk_private_signals + signal_id : \
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
(GtkSignal*) 0 \
)
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static GtkSignalMarshal global_marshaller = NULL;
static GtkSignalDestroy global_destroy_notify = NULL;
1997-11-24 22:37:52 +00:00
static guint gtk_handler_id = 1;
static guint gtk_handler_quark = 0;
static GHashTable *gtk_signal_hash_table = NULL;
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
GtkSignal *_gtk_private_signals = NULL;
guint _gtk_private_n_signals = 0;
static GMemChunk *gtk_signal_hash_mem_chunk = NULL;
static GMemChunk *gtk_disconnect_info_mem_chunk = NULL;
static GtkHandler *gtk_handler_free_list = NULL;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
static GtkEmission *gtk_free_emissions = NULL;
1997-11-24 22:37:52 +00:00
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
static GtkEmission *current_emissions = NULL;
static GtkEmission *stop_emissions = NULL;
static GtkEmission *restart_emissions = NULL;
1997-11-24 22:37:52 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static GtkSignal*
gtk_signal_next_and_invalidate (void)
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
static guint gtk_n_free_signals = 0;
register GtkSignal *signal;
register guint new_signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
/* don't keep *any* GtkSignal pointers across invokation of this function!!!
*/
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (gtk_n_free_signals == 0)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
register guint i;
register guint size;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
/* nearest pow
*/
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
size = _gtk_private_n_signals + SIGNAL_BLOCK_SIZE;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
size *= sizeof (GtkSignal);
i = 1;
while (i < size)
i <<= 1;
size = i;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
_gtk_private_signals = g_realloc (_gtk_private_signals, size);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
gtk_n_free_signals = size / sizeof (GtkSignal) - _gtk_private_n_signals;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
memset (_gtk_private_signals + _gtk_private_n_signals, 0, gtk_n_free_signals * sizeof (GtkSignal));
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
new_signal_id = _gtk_private_n_signals++;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_n_free_signals--;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
Adapt cast macros to standard. Fri Nov 19 10:34:41 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkgamma.h: Adapt cast macros to standard. [ Merges from 1.2 ] Tue Nov 16 10:15:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkitemfactory.c (gtk_item_factory_parse_path): If translation does not include a '/', use entire translation instead of crashing. Sun Oct 31 22:21:11 1999 Tim Janik <timj@gtk.org> * docs/gtk_tut.sgml: s/gtk_accel_group_attach/gtk_window_add_accel_group/. Sat Oct 30 09:09:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c (gtk_target_list_remove): Use g_list_remove_link, not g_list_remove. [ From Geert Bevin <gbevin@thunderstorms.org> ] Sun Oct 24 07:41:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkclist.c (real_undo_selection): commented out g_print() statement upon unselection (how the heck did that slip in?). Sat Oct 23 03:03:08 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_freeze): (gtk_text_thaw): undraw/draw cursor here to avoid unnecessary scrolling in frozen state (and aparently crashes). patch provided by Anders Melchiorsen <and@kampsax.dtu.dk>. Sat Oct 23 02:53:20 1999 Tim Janik <timj@gtk.org> * fix insensitive default/focus widget activation, reported by Matt Goodall <mgg@isotek.co.uk>. * gtk/gtkwindow.c (gtk_window_key_press_event): (gtk_window_activate_default): (gtk_window_activate_focus): return handled=FALSE for actiavtion of insensitive default widgets. return handled=TRUE for activation of insensitive focus widgets. don't activate in either case. Tue Oct 19 09:55:08 1999 Owen Taylor <otaylor@redhat.com> * gtk-config.in (lib_gtk): Switch order of @x_cflags@ and $glib_cflags to match library order and in the theory that an old version of GLib is more likely to be in the include directory for X then vice-versa. (Bug #2776) Tue Oct 19 09:46:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always use LC_CTYPE to determine the locale for fontsets, not LC_MESSAGES; the user may want English messages with a handling for non-English languages. (LC_CTYPE=ru_RU LC_MESSAGES=fr_FR will still be broken) (Bug #2891) Tue Oct 19 20:36:42 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Fixed the wrong number of "*"'s. Tue Oct 19 12:15:13 1999 Changwoo Ryu <cwryu@adam.kaist.ac.kr> * gtk/gtkrc.ko: Changed the Korean default fontset. Fri Oct 8 02:32:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtklayout.c (gtk_layout_adjustment_changed): Fix bug where when scrolling to the left or top double exposes were done, causing major slowdowns. Thu Oct 7 18:31:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Always normalize codeset names to lowercase and alphanumeric, before looking them up. * gtk/Makefile.am: Install codeset variant gtkrc files with normalized names. Thu Oct 7 22:52:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_popup_with_data): even popup menus when the menu is already visible, but its parent is still hidden, (happens after tornoff window got hidden). Thu Oct 7 11:09:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (install-data-local): Fix a typo where gtkrc.vi_VN.tcvn5712 wasn't getting deleted, causing error messages on install. Thu Oct 7 11:03:06 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtk[hv]paned.c (gtk_{h,v}paned_draw): Redraw the handle as well, since we now sometimes ignore exposes on the handle while resizing. Wed Oct 6 18:02:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_append_default_pixmap_path): Fix stupid extra g_free introduced in one of the last one or two commits. Wed Oct 6 16:38:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc*: Make naming of gtkrc files consistent, put each style that a gtkrc.* file creates in a unique namespace, remove old files before installing. Wed Oct 6 14:31:16 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_init): Avoid leaking memory when gtk_rc_init is called multiple times. (Yes, people who do that have bugs in their code.) Tue Oct 5 11:36:57 PDT 1999 Manish Singh <yosh@gimp.org> * gtk/Makefile.am: listing gtkrc in gtkconf_DATA seems to barf. Remove it since it's generated by the Makefile anyway Tue Oct 5 02:43:41 1999 Owen Taylor <otaylor@redhat.com> * gdk/gdkinputcommon.h (gdk_input_device_new): Free device->info.axes for core pointer. Thu Sep 30 13:55:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (destroy_idle_test): Rename idle to idle_id, to deal with obsolete, broken C libraries. Mon Sep 27 02:50:15 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkvscale.c (gtk_vscale_draw): Don't add in allocation->x/y twice! * gtk/gtkhscale.c: Make usage of gtk_hscale_pos_trough() consistent with gtk_vscale_pos_trough(). Sun Sep 26 19:44:34 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcontainer.c (gtk_container_queue_resize): We may be queueing a resize on a toplevel container between the time we show it and when we map it. So, we need to test GTK_WIDGET_VISIBLE() for toplevels, and only use GTK_WIDGET_DRAWABLE() for child windows. Thu Sep 23 16:41:03 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am (LDADDS): Add GDK_WLIBS to LDADDS (fixes bug #2144) Fri Sep 24 00:51:45 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c (gtk_container_queue_resize): check for container DRAWABLE (instead of VISIBLE), so we don't queue resizes on non-toplevel containers. Sun Sep 19 18:13:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.[ch]: Make the data argument const guchar *. Sat Sep 18 21:27:40 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c: Try to behave sensibly if the focus widget is the window itself. (Should we allow this at all?) Fri Sep 17 09:57:15 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_set_sensitive): * gtk/gtknotebook.c (gtk_notebook_set_scrollable): * gtk/gtknotebook.c (gtk_notebook_set_show_border): * gtk/gtkclist.c (gtk_notebook_set_show_border): make gboolean args in prototypes and implementations consistent (Tomas Ogren). * gtk/gtklayout.c (gtk_layout_remove): unset GTK_IS_OFFSCREEN flag before the widget is unparented (reported by damon). * gtk/gtkdnd.c: make the cursor and icon data _unsigned_ char, since we provide unsigned data anyways. Thu Sep 16 21:32:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktooltips.c (gtk_tooltips_set_tip): Delay the call to gtk_tooltips_layout_text() until later. * gtk/gtktooltips.c (gtk_tooltips_draw_tips): Call gtk_widget_ensure_style() before using the style. Wed Sep 15 02:52:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_event): Added workaround for old widgets that don't propagate draws to all children. (Namely gnome-dock for gnome-libs <= 1.0.16) Tue Sep 14 19:22:19 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_queue_resize): don't queue the parent for a redraw but just the widget that requested the resize. Tue Sep 14 18:29:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: added new widget level method gtk_container_set_reallocate_redraws() and a GtkContainer flag reallocate_redraws : 1 to reflect the setting, exported this through the argument system as a boolean ::reallocate_redraws. * gtk/gtkwidget.c (gtk_widget_size_allocate): when queueing redraws on the widget because the allocation changes, do so as well for widget->parent if the parent has reallocate_redraws set to TRUE. with that containers requesting reallocation redraws get automatically redrawn if their children changed allocation (this unfortunately affects also other children that didn't change allocation, but we cannot work around that before 1.3). Tue Sep 14 18:23:01 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_idle_draw): only emit ::draw if width _and_ height are >0 (not _or_). * gtk/gtktable.c (gtk_table_remove): use gtk_widget_queue_resize() instead of gtk_container_queue_resize(), which is a core gtk internal function (must have been on crack when i queued that). * gtk/gtkprivate.h: added new private flag GTK_FULLDRAW_PENDING, so we can check more reliably if we want to discard expose events. * gtk/gtkwidget.c: added setting/unsetting of the GTK_FULLDRAW_PENDING flag. * gtk/gtkwidget.c (gtk_widget_event): don't discard synthesized exposes, we simply trust these events. for deciding whether to discard exposes, check GTK_FULLDRAW_PENDING instead of RESIZE_PENDING. Mon Sep 13 15:01:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_queue_clear_area): Ignore queues of areas that are completely off screen. * gtk/gtkwidget.c (gtk_widget_idle_draw): Fix broken logic for handleboxes. * gtk/gtkwidget.c (gtk_widget_queue_draw_data): Add santity check on width/height. Mon Sep 13 02:22:47 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: backed out Raja's recent VPATH build "improvements", we are fine with using $@ the way we do (if we actually encounter brokeness with $@ in VPATH builds because of additional path prefixes, we need to use $(@F) actually). Mon Sep 13 01:34:53 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: renamed two variables, hopefully didn't introduce short lived bugs, that would allow language bindings to do surgeries to our guts. Fri Sep 10 15:22:50 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_size_allocate): Fix typo where comparison was being done against an uninitialized value causing intermittant results depending on compiler flags. Also make it clearer that we aren't ever initializing the child as 0x0 (though this will be caught in gtk_widget_size_allocate()) Fri Sep 10 10:06:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_transient_for): Move gtk_window_unset_transient_for() call after we do checks involving the old transient parent. [ From Lance Capser <lmc@cyberhighway.net> ] 1999-09-07 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gtk.defs): Go back to using `touch' to create an empty file. Mon Sep 6 00:11:56 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (gtk_ctree_is_ancestor): return FALSE if node has no children. (Reported by: Chris Rogers <gandalf@pobox.com>) 1999-09-03 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (gen_sources): Improve VPATH builds. `$@' is valid only in the build dir, not after we've done `cd $srcdir'. Also use `test -f' instead of less portable `test -e'. Fri Sep 3 15:59:56 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (correct_cache_insert): Rewrite for simplicity, and hopefully correctness. (Fixes bug #1322, which was a segfault when on some insertions with the properties around the insertion set up just wrong.) * gtk/gtktext.c (gtk_text_adjustment): When we receive a "changed" signal, clamp the new value to the adjustment bounds to avoid segfaulting if someone tries to change the adjustment to a bogus value. (Bug #1795) Thu Sep 2 16:33:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c: Ignore unexpected destroy notifies for children, for toplevel windows handle them like delete_event. * gtk/gtkplug.c: Add an unrealize handler so that we unref plug->socket_window when we are done with it. Fri Sep 3 14:52:54 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c (clear_area): Fix stupid signedness problem that was causing background to sometimes be misaligned. Fri Sep 3 12:26:33 1999 Owen Taylor <otaylor@redhat.com> [ Fixes pointed out by Ettore Perazzoli <ettore@comm2000.it> ] * gtk/gtkmenu.c (gtk_menu_position): Make sure we never position menus with negative x, y, since gtk_widget_set_uposition() can't handle that. * gtk/gtkmenuitem.c (gtk_menu_item_position_menu): Modify the positioning code a bit so that we always put the top-left corner onscreen. (This is for UI reasons, gtk_menu_position() now takes care of gtk_widet_set_uposition() brokeness.) Fri Sep 3 03:06:30 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: fixed up things for -jx, x > 1. Sun Sep 5 08:48:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_modify_style): Ref the RC style that is passed in. The lack of the ref before was a bug. If people worked around this bug, this will introduce a slight memory leak in their code. The code should typically look like: rc_style = gtk_rc_style_new (); [...] gtk_widget_modify_style (widget, rc_style); gtk_rc_style_unref (rc_style); * gtk/gtkwidget.c (gtk_widget_modify_style): Reset the style if it was already set. * gtk/gtkwidget.c (gtk_widget_set_name): Only set the style if it was set before. Thu Sep 2 19:02:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (main): Add a check to see if we are being run from the correct directory and to quit nicely if we are not. * gtk/gtkrc.c (gtk_rc_slist_remove_all): Make function static. Thu Sep 2 23:00:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkenums.h (GtkWindowPosition): added GTK_WIN_POS_CENTER_ALWAYS. * gtk/gtkwindow.c: queue resizes unconditionally (gtk_widget_queue_resize will figure what to do if the window is not realized). (gtk_window_move_resize): only recenter the window for GTK_WIN_POS_CENTER_ALWAYS. (gtk_window_compute_reposition): handle GTK_WIN_POS_CENTER_ALWAYS in the same way as GTK_WIN_POS_CENTER. Thu Sep 2 22:39:27 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_reposition): check for the last position with (!(info->last_flags & GDK_HINT_POS)) instead of (!info->last_flags & GDK_HINT_POS). * gtk/gtkwindow.c (gtk_window_move_resize): constrain new_width and new_height unconditionally, because we use these values even if !default_size_changed && !hints_changed. comented the (default_size_changed || hints_changed) case with respect to resize rejects from the window manager. * gtk/gtkwindow.c (gtk_window_move_resize): save info->last values in the zvt condition hack, since this includes the window hints, set the hints after the handling_resize case. * gtk/gtkwindow.c (gtk_window_show): constrain the default size that a window is initially shown with to the geometry. Thu Sep 2 07:38:56 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): s/size_changed/default_size_changed/g so i know what's really going on (frying brain on smaller flame now). Thu Sep 2 05:47:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): queue a new resize if we have size_changed upon handling_resize. this is a gross workaround for the broken zvt widget and should be removed in 1.3 again (search for FIXME). Owen provided an accurate comment for this: /* We could be here for two reasons * 1) We coincidentally got a resize while handling * another resize. * 2) Our computation of size_changed was completely * screwed up, probably because one of our children * is broken. It's probably a zvt widget. * * For 1), we could just go ahead and ask for the * new size right now, but doing that for 2) * might well be fighting the user (and can even * trigger a loop). Since we really don't want to * do that, we requeue a resize in hopes that * by the time it gets handled, the child has seen * the light and is willing to go along with the * new size. (this happens for the zvt widget, since * the size_allocate() above will have stored the * requisition corresponding to the new size in the * zvt widget) * * This doesn't buy us anything for 1), but it shouldn't * hurt us too badly, since it is what would have * happened if we had gotten the configure event before * the new size had been set. */ Wed Sep 1 20:46:11 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c: deal properly with the fact that RC style lists may include rc styles more than once. * gtk/gtkrc.c (gtk_rc_append_pixmap_path): Removed unused static function. * gtk/gtkrc.c (gtk_rc_find_pixmap_in_path): Keep a stack of directories of RC files currently being parsed and implicitely add them to pixmap path. This fixes a bug where the directory would get appended then overwritten by pixmap_path declarations. (bug #1462, from Peter Wainright <prw@wainpr.demon.co.uk>) * gtk/gtkthemes.c (gtk_theme_engine_unref): Call theme's exit function. (Patch from Peter Wainwright, bug #1454) * gtk/gtkradiomenuitem.c (gtk_radio_menu_item_destroy): Add a destroy() handler to take care of removing group for menu item. (Fixes bug #1197) * gtk/gtkwidget.c (gtk_widget_size_request): Fixed thinko in warning message. Wed Sep 1 21:27:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_move_resize): don't require a server roundtrip to figure window's width and height, since we know that anyways from widget->allocation. Wed Sep 1 12:37:44 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_move_resize): Compute the hints after we request the new size. Wed Sep 1 10:38:37 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_compute_hints): Removed GTK_WIDGET_REALIZED() assertion - we can compute the hints before we are realized. * gtk/gtkwindow.c (gtk_window_move_resize): Reorder hint changing so that we have a value of hints_changed when we decide whether to constrain the window size. * gtk/gtkwindow.c (gtk_window_move_resize): Spelling fix. * gtk/gtkwindow.c (gtk_window_constrain_size): cleanups, change back to G_MAXINT. Wed Sep 1 06:54:59 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c (gtk_window_get_geometry_info): zero initialize new GtkWindowGeometryInfo, so fields like GdkGeometry geometry contain uncluttered values. (gtk_window_compute_hints): simply assert that window is realized and that geometry_info is valid, since we rely on this anyways. (gtk_window_constrain_size): major cleanups to the code. if (flags & GDK_HINT_BASE_SIZE) use geometry's base width and height for the base size, instead of the minimums. use 32767 as max width and height (like in gtkwindow.c) instead of G_MAXINT. Wed Sep 1 04:41:25 1999 Tim Janik <timj@gtk.org> * cleaned up the GtkContainer.need_resize flag handling mess, we only need to force resize requests when we were prematurely realized, or our widget tree was modified when we were temporarily hidden. handling these cases directly upon showing the window (i.e. while the GdkWindow is still unmapped) avoids the need to wait for a configure event response and therefore makes the GUI more snappier and avoids blank windows during the roundtrip. * gtk/gtkwidget.c: (gtk_widget_hide): (gtk_widget_show): don't queue resizes on toplevels, they know how to deal with matters. * gtk/gtkcontainer.c (gtk_container_queue_resize): set the ->need_resize flag directly for not visible resize containers and spare us unecessary signal emissions. * gtk/gtkwindow.c: (gtk_window_realize): if we need to enforce premature size allocation, queue a container resize so we are correctly resized later on. (gtk_window_init): (gtk_window_size_request): don't freak around with the ->need_resize flag, gtk_container_queue_resize() will care about that. (gtk_window_show): handle initial resizing issues here, we can handle matters better in this place, especially since we know that our GdkWindow is still unmapped. (gtk_window_move_resize): don't care about ->need_resize at all. handle size changes properly that occoured while we waited for a configure event. Tue Aug 31 15:58:46 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_set_uposition): * gtk/gtkwindow.[hc] (gtk_window_reposition): Move the hint setting code from gtk_widget_set_uposition to here; set the hints so that we respect any previously set geometry hints. * gtk/gtkwindow.c (gtk_window_compute_reposition): Don't change the window hints here or move the window here, let that happen in gtk_window_move_resize(). Tue Aug 31 06:58:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): when discarding exposes due to already queued resizes (and therefore redraws) on a widget, check its anchestry as well. * gtk/gtkcontainer.c: (gtk_container_queue_resize): clear resize widgets for resize containers before aborting prematurely. this is especially important for toplevels which may need imemdiate processing or their resize handler to be queued. (gtk_container_dequeue_resize_handler): added new internal function for gtkwindow.c. * gtk/gtkwindow.c (gtk_window_move_resize): if we are resizing due to a configure event, take possible changes in window position into account as well. if we request a new window size, queue up a resize handler that will last until the configure event response arrives. combined the ->need_resize case (initial show) with the general size (hints) changed case and added even more comments. if !auto_shrink, only revert to the old allocation if the new size is smaller than the current allocation. Tue Aug 31 11:55:20 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c (gtk_window_set_arg, gtk_window_set_policy, gtk_window_set_geometry_hints gtk_window_set_default_size): When hints are set, queue a resize so that the hints will be eventually reset on the toplevel. * gtk/gtkwindow.c (gtk_window_show): Use gtk_window_compute_default_size(). Clear the need_resize flag on the initail map so that we don't unnecessarily trigger the resize code. * gtk/gtkwindow.c (gtk_window_move_resize): Split apart into separate functions. Compare the hints we are setting with what we set last time so that we can accurately tell when we need to reset the hints. * gtk/gtkwindow.c (gtk_window_compute_default_size): New function to figure out the size from requisition and default_size. * gtk/gtkwindow.c (gtk_window_constrain_size): Function from fvwm to constrain a size to the geometry hints. * gtk/gtkwindow.c (gtk_window_compare_hints): New function to compare two sets of geometry hints. * gtk/gtkwindow.c (gtk_window_compute_hints): Renamed from gtk_window_set_hints(), just compute the hints, don't set them. * gtk/gtkwindow.c (gtk_window_compute_reposition): Move code from gtk_window_move_resize() to separate function, rationalize a bit. Tue Aug 31 13:05:03 1999 Owen Taylor <otaylor@redhat.com> * gtkrc.h: Move the ref_count member out of the GtkRcStyle structure into a new private structure. * gtkrc.c: Split GtkRcStyle into public/private. In the private part, add a list of pointers to the RcStyle lists this RcStyle participates in. * gtkrc.c: When a RcStyle is free, remove all lists referencing it from the realized_style_ht hash, and free those lists. * gtk/gtkrc.c (gtk_rc_clear_styles): Don't call gtk_rc_init(), since that adds the default styles to the list of parsed RC files again. * gtk/gtkrc.c: Use gtk_rc_style_find() consistently. Thu Aug 26 14:14:42 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): reactivated the sanity checks that ensure that a widget's allocation is at least 1 in width and height. (GNOME note: this doesn't affect old panel code anymore, because GtkSocket will request width and height of at least 1 since Fri Jul 23). * gtk/gtkwindow.c (gtk_window_realize): if the widget hasn't been allocated yet (happens if the user realizes the window prematurely), size request and allocate it. (gtk_window_size_allocate): guard against guint underflows. Wed Aug 25 19:01:36 1999 Lars Hamann <lars@gtk.org> * gtk/Makefile.am (gtk_built_sources): reordered stamp-gtk.defs in gtk_built_sources, so gtk.defs gets built prior to all other sources. Mon Aug 23 19:11:17 1999 Tim Janik <timj@gtk.org> * gtk/Makefile.am: invoke indent on gtkmarshal.*. rewrote source generation rules, use COPYING as oldest source tag for a piggyback rule to generate all sources from (don't touch it ;). major cleanups, strip spaces on build rules for GNU Make. * gtk/genmarshal.pl: don't operate on hardcoded filenames but take source and target files from commandline arguments. don't invoke indent. Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Sat Aug 21 14:07:36 1999 Lars Hamann <lars@gtk.org> * gtk/gtknotebook.c (gtk_notebook_real_switch_page): queue_resize notebook after switch to avoid drawing problems. (Bug#856 Reported by: Theodore Roth <troth@netmagic.net>) (gtk_notebook_pages_allocate): don't map not visible tab_labels, show them instead. (Bug#1805 Reported by: Dave Cole <dave@dccs.com.au>) Wed Aug 18 09:20:10 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: we use window->need_resize from configure_event now, to indicate that the gtkwindow should keep its allocation (e.g. because the user resized the window through window manager handles). resize_count is now reliably used to figure whether we got the allocation we requested from the window manager. configure events get queued as resizes now, the real stuff (size computation and allocation) now only goes on in gtk_window_move_resize(). GtkWindow's requisition now contains its *real* requisition (like all other widgets), *not* taking usize into account. geometry_info->last_{width|height} is now updated from set_hints() only so it always contains the last hints we set for the window manager. made some event handlers return TRUE instead of FALSE. the overall code should be much more straight forward now, and the significant code portions are accompanied by comments now. (gtk_window_set_hints): removed requisition argument and made it fetch the requisition through gtk_widget_get_child_requisition. we also don't move the gdkwindow here anymore, gtk_window_move_resize() does that now. (gtk_window_show): ensure that the widget is realized before calling gtk_container_check_resize() (and thus gtk_window_move_resize()), also ensure that we got properly size requested and allocated before realization. (gtk_window_configure_event): ignore plain window moves, or reallocate the widget tree through the resize queue otherwise. (gtk_window_move_resize): mostly rewrote this function to figure window manager hints more reliably, coalesce window moves and resizes to reduce configure events and do actuall size allocations. Tue Aug 17 07:43:04 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_event): discard expose events for widgets that have a resize pending, because a redraw is already queued for them. * gtk/gtkcontainer.c: removed ugly connect_to ::size_allocate signal hack to clear resize_widgets. * gtk/gtkwidget.c (gtk_widget_size_allocate): call clear_resize_widgets for resize containers prior to size allocation. (this is also a bit ugly, but avoids side effects for stopped emissions and is thus more reliable). (gtk_widget_unparent): removed disconnect call for clear_resize_widgets. * gtk/gtktooltips.c (gtk_tooltips_paint_window): renamed this function from gtk_tooltips_expose, as we connect to ::expose_event *and* ::draw now. 1999-08-18 Federico Mena Quintero <federico@redhat.com> * gtk/gtkselection.c (gtk_target_list_ref): Added missing sanity checks. (gtk_target_list_unref): Likewise. * gtk/gtkthemes.c (gtk_theme_engine_unref): Likewise. Tue Aug 17 15:47:07 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c (gtk_color_selection_draw_value_bar): guard against division by zero. (Fixes bug #1339) Tue Aug 17 10:56:49 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_move_{forward,backward}_word): Prevent the trivial leak of information of allowing word motion when the entry is not visible. Tue Aug 17 10:28:52 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (gtk_file_selection_fileop_error): Propagate modality to error dialog as well as confirmation dialogs. (Bug #1803, reported by Rosanna Wing Sze Yuen) Wed Aug 11 01:04:57 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktreeitem.c (gtk_tree_item_draw_lines): Honor tree->view_lines. (gtk-guy-990611-3.patch: Guy Harris <guy@netapp.com>) * gtk/Makefile.am (install-data-local): Solaris apparently has various troubles with ln -f; use rm first instead. (gtk-guy-990611-2.patch: Guy Harris <guy@netapp.com>) 1999-07-30 Raja R Harinath <harinath@cs.umn.edu> * gtk/Makefile.am (stamp-m): Don't `mv' from builddir to srcdir, use `cp' followed by `rm' (the `rm' was already there). July 30, 1999 Elliot Lee <sopwith@redhat.com> * configure.in: Fix autoconf warnings about cross compilation by trying to provide sane defaults for AC_TRY_RUN. * gtk/Makefile.am: If we refer to gtkmarshal.[ch] in $(srcdir), put them into $(srcdir) when generated. Also add a dependency of gtksignal.h on gtkmarshal.h for -j builds. * gtk/gtk(dnd,style,gamma).c: Minor warning fixes. Wed Jul 28 09:29:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration_do): Added missing GDK_THREADS_{LEAVE,ENTER} pair. (From Paul Fisher <pnfisher@redhat.com>) Fri Jul 23 01:00:15 1999 Tim Janik <timj@gtk.org> * gtk/gtksocket.c (gtk_socket_size_request): asure that the requested width and height are always >0 (owen). Fri Jul 23 00:00:47 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): backed out my recent change that assured that a widget's allocated with and height are always >1, since this breaks *buggy* panel code. unfortunately this back-breaks the gimp's color selector. * gtk/gtkdrawingarea.c (gtk_drawing_area_size_allocate): asure that our allocation is always >0 in width and height, before sending the configure event; this is a *gross* hack to get the gimp back to work. * marked both cases with TODO-1.3 Wed Jul 21 15:47:39 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtktext.c: Don't display wrap indicators when text is not editable and word wrap is on. Wed Jul 21 08:21:40 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_create_item): special case option menus here as they are not derived from menu shell, assure that the option menu has a menu we can add items to. Tue Jul 20 23:29:48 1999 Tim Janik <timj@gtk.org> * gtk/gtknotebook.c (gtk_notebook_page_allocate): convert allocation-> width/height to (gint) before calculations and check against < 0 to avoid guint wraparounds. Sun Jul 18 00:35:49 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_size_allocate): ensure that the allocated width and height is never zero. sanity check both dimensions against 32767 and issue a warning if the allocation is greater than that. Wed Jul 7 15:03:30 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_events_pending): Unlock around call to g_main_pending() as well. Wed Jul 7 14:59:01 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main_iteration): Unlock around call to g_main_iteration() - since that will regrab GTK+ lock to process events. Thu Jul 1 15:01:55 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkwindow.c - Regularize with the rest of GTK+ by making widget->requisition not reflect the set_usize() - Always recompute geometry hints, then check if they changed before sending them to the X server. The previous checks for changes would fail in a number of circumstances. Thu Jul 1 11:55:59 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Include <stdlib.h> for strcmp(). Wed Jun 30 19:26:36 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c: - Code cleanups - Instantaneously update on modifier key presses - Allow cancellation of the drag with Escape. Tue Jun 29 17:04:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/testgtk.c (create_handle_box): Set the policy to auto_shrink - otherwise the appearance is rather strange when flipping between horizontal and vertical. Mon Jun 28 09:29:52 1999 Tim Janik <timj@gtk.org> * gtk/gtkwindow.c: add ::default_width and ::default_height arguments. (gtk_window_set_default_size): don't change a value if it's < 0. queue a resize. Sun Jun 27 11:00:33 1999 Tim Janik <timj@gtk.org> * gtk/gtktext.c (gtk_text_insert): don't segfault on NULL inserts. Mon Jun 28 12:08:25 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkfilesel.c (cmpl_completion_fullname): Don't add an extra "/" when concating "/" + filename. (From Matt Grossman <mattg@oz.net>) Mon Jun 28 10:57:12 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkctree.c (draw_cell_pixmap): Reset clip mask for fg_gc if we set it for drawing pixmap. Tue Jun 15 12:45:12 1999 Owen Taylor <otaylor@redhat.com> Fixes from Peter Wainwright <prw@wainpr.demon.co.uk> * gtk/gtkrc.c (gtk_rc_parse_engine): If rc_style->engine is already set, call old engine's destroy function and unref the old engine. Thu Jun 10 17:59:38 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.c (gtk_drawing_area_size): queue a resize. Wed Jun 9 15:13:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: mark certain functions as internal. Wed Jun 9 13:48:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkpreview.c (gtk_preview_set_expand): queue a resize if the expand behaviour changed. * gtk/gtklabel.c (gtk_label_set_pattern): (gtk_label_set_justify): (gtk_label_set_line_wrap): don't bother invoking queue_clear, the reallocation does that for us, always free_words so the upcoming resize will relayout the label's contents. Wed Jun 9 12:50:48 1999 Tim Janik <timj@gtk.org> * applied argument implementation patches from Elena Devdariani <elena@cogent.ca>. * gtk/gtktoolbar.c: ::orientation, ::toolbar_style, ::space_size, ::space_style, ::relief * gtk/gtkruler.c: ::lower, ::upper, ::position, ::max_size * gtk/gtkpreview.c: ::expand * gtk/gtkpaned.c: ::handle_size, ::gutter_size * gtk/gtknotebook.c: ::homogeneous * gtk/gtklabel.c: ::wrap * gtk/gtklist.c: ::selection_mode * gtk/gtkhandlebox.c: ::handle_position, ::snap_edge * gtk/gtkcurve.c: ::curve_type, ::min_x, ::max_x, ::min_y, ::max_y * gtk/gtkcolorsel.c: ::update_policy, ::use_opacity * gtk/gtkclist.c: ::sort_type * gtk/gtkcheckmenuitem.c: ::active, ::show_toggle * gtk/gtkaspectframe.c: ::xalign, ::yalign, ::ratio, ::obey_child Tue Jun 1 23:38:38 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c: Removed ill-thought-out part of last comment. Tue Jun 1 23:30:09 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkstyle.c (gtk_style_attach): Get the reference counting right when we have to attach a new style for a different visual. (Chi-Deok Hwang <cdhwang@sr.hei.co.kr>) * gtk/gtkstyle.c: Documented the refcounting peculularities of gtk_style_attach. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkwindow.c (gtk_window_new): added return if fail for invalid window types. * examples/packer/pack.c (main): use GTK_WINDOW_TOPLEVEL instead of GTK_TOPLEVEL for creating the window. 1999-06-01 Tim Janik <timj@dhcpd7.redhat.com> * gtk/gtkmain.c (gtk_init_check): don't segfault when --gtk-module is the last argument (reported by Per Winkvist). Tue May 25 13:13:12 1999 Owen Taylor <otaylor@redhat.com> Fixes for invisible XOR lines (Frank Loemker <floemker@TechFak.Uni-Bielefeld.DE>) * gtk/gtkclist.c (gtk_clist_realize): Always use a non-zero pixel for GDK_XOR. * gtk/gtkvpaned.c gtk/gtkhpaned.c: Use GDK_INVERT instead of GDK_XOR. Wed May 12 21:56:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (adjust_adjustments): signal emit value_changed signals if h/voffsets differ from adjustment values. Reportet by Jerome Bolliet <bolliet@in2p3.fr> Mon May 10 04:20:41 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): propagate ::selection-done emissions up to the topmost menu shell. Fri May 7 10:15:14 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtknotebook.c (gtk_notebook_set_scrollable): Unset the user data on the window before destroying it. * gtk/gtknotebook.c (gtk_notebook_unrealize): Add an unrealize handler to take care of destroying notebook->panel properly. (Bug #1198 - Morten Welinder <terra@diku.dk>) * gtk/gtktext.c (expand_scratch_buffer): Fix reversal of g_new and g_realloc to stop memory leak. (Actually, we could just use g_realloc(), but I'm not 100% sure that is portable). (Bug #1196 - Morten Welinder <terra@diku.dk>) Wed Apr 21 00:42:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkplug.h: Removed stray GtkPlugButton declaration. * gdk/gdkfont.c (gdk_text_measure): Fix the return value for fontsets. * gtk/gtkbutton.c (gtkbutton_expose): Fix warning with bin/button confusion. Thu May 6 04:53:26 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c: in the ItemFactory test, link radio items together, and show how preselection of radio items is done. Sun May 2 13:31:14 1999 Tim Janik <timj@gtk.org> * gtk/gtktreeitem.c (gtk_tree_item_set_subtree): * gtk/gtktree.c (gtk_tree_add) (gtk_tree_insert): * gtk/gtktoolbar.c (gtk_toolbar_insert_element): * gtk/gtkpaned.c (gtk_paned_pack2) (gtk_paned_pack1): * gtk/gtkscrolledwindow.c (gtk_scrolled_window_add): * gtk/gtktable.c (gtk_table_attach): * gtk/gtklist.c (gtk_list_insert_items): * gtk/gtkmenushell.c (gtk_menu_shell_insert): * gtk/gtknotebook.c (gtk_notebook_insert_page_menu): * gtk/gtkpacker.c (gtk_packer_add_defaults) (gtk_packer_add): * gtk/gtkbin.c (gtk_bin_add): * gtk/gtkbox.c (gtk_box_pack_start) (gtk_box_pack_end): * gtk/gtkfixed.c (gtk_fixed_put): * gtk/gtklayout.c (gtk_layout_put): general fixups to container_add logic. always realize child if child->parent is realized, only map the child and queue a resize if child and child->parent are both visible. Fri Apr 30 09:02:28 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_real_unrealize): use gtk_container_forall instead of gtk_container_foreach to walk and unrealize children, so composite children get also unrealized. (gtk_widget_real_show): don't call gtk_widget_map() if we don't need to. (gtk_widget_map): assert that the widget is visible (basic constrain). (gtk_widget_real_map): assert that the widget is realized (basic constrain). Fri Apr 29 00:53:20 1999 Tim Janik <timj@gtk.org> * gtk/gtkbindings.c (gtk_pattern_spec_init): plugged a memory leak. Tue May 4 09:32:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am gtk/gtkrc.iso-8859-2: Add a gtkrc file for iso-8859-2 locales. * configure.in (ALL_LINGUAS): ALL_LINGUAS update. Tue Apr 27 16:38:32 1999 Owen Taylor <otaylor@redhat.com> * gtk/Makefile.am: Fix typo of static_sources for static_SOURCES. (Pointed out by andy@rz.uni-karlsruhe.de and others). Remove some suspicious and useless lines. Tue May 4 08:44:08 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkclist.c (gtk_clist_draw, draw_rows): Use width,height = 0, 0 to mean - here to edge of window, instead of -1, -1, since the former is all we support. Tue May 4 08:34:43 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkmain.c (gtk_main): Correctly free list nodes when removing from quit_functions list. Tue Apr 27 14:17:16 1999 Tim Janik <timj@gtk.org> * gtk/gtkpacker.c (gtk_packer_size_request): remove unused variable. Tue Apr 27 18:23:35 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkentry.c (gtk_entry_draw_cursor_on_drawable): When redrawing characters on non-visible entry, use appropriate '*' character. (Bug #1130 - Jean-Marc Jacquet <jm@littleigloo.org>) Tue Apr 27 01:31:40 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (drag_dest_cell): compute destination cell from drag coordinates. (gtk_clist_drag_data_received) (gtk_clist_drag_motion): use drag_dest_cell. * gtk/gtktree (drag_dest_cell) (gtk_ctree_drag_data_received) (gtk_ctree_drag_motion): likewise. (Bug #1129) Wed Apr 21 21:26:11 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_init): properly initialize translate_* fields. (gtk_item_factory_finalize): invoke translate_notify independant from translate_data. (gtk_item_factory_set_translate_func): likewise. (gtk_item_factory_destroy): only remove ifactory pointer from those widgets that belong to us (stupid me). Mon Apr 19 12:05:31 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkrc.c (gtk_rc_style_init): Fixed leak of rc_style list when lookup succeeeds. Thu Apr 15 01:11:24 1999 Lars Hamann <lars@gtk.org> * gtk/gtkctree.c (resync_selection): * gtk/gtkclist.c (resync_selection): fixed undo_selection bug. * gtk/gtkclist.c (gtk_clist_button_release): fixed resync_selection bug. Fri Apr 9 19:22:19 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkcolorsel.c gtk/gtkdnd.c gtk/gtkmenuitem.c: Add some missing GDK_THREADS_ENTER()/LEAVE around timeouts. (Patches from Sebastian Wilhelmi <wilhelmi@ira.uka.de>) Thu Apr 8 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkentry.c (gtk_entry_key_press): use gtk_widget_activate() rather than emit_by_name. * gtk/gtkeditable.c (gtk_editable_insert_text): keep a reference on the widget across multiple signal emissions. (gtk_editable_delete_text): same here. (gtk_editable_class_init): set widget_class->activate_signal after editable_signals[ACTIVATE] has been created. Wed Apr 7 22:59:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkpaned.c (gtk_paned_set_position): Don't clamp position here prematurely -- we might not have the right ->min_position and ->max_position yet. Tue Apr 6 16:38:51 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkselection.c gdk/gdkselection.c: (gtk_selection_request): Add error traps so if the other end of the connection dies, we survive. * gtk/gtkselection.c (gtk_selection_notify): Clean up properly when selection property retrieval fails. * gtk/gtkselection.c (gtk_selection_request): Correctly reject SelectionRequest notifies where the handler returns no data. Tue Apr 6 12:24:21 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkdnd.c (gtk_drag_dest_leave): Only unhighlight when we've previously highlighted. * gtk/gtkdnd.c (gtk_drag_dest_handle_event): Don't emit two "drag_leave" signals for Motif drops. * gtk/gtkdnd.c (gtk_drag_source_handle_event): Send back the correct status messages when dropping from Motif onto a proxy window that is rejecting the drop. Sat Mar 27 23:32:13 1999 Tim Janik <timj@gtk.org> * gtk/gtkarg.[hc]: added functions from GLE, gtk_arg_reset() to free the value and reset type to GTK_TYPE_INVALID, and gtk_arg_values_equal() to compare two argument values. added gtk_arg_to_valueloc() to set a variable from an arg through its location (pointer). * gtk/gtkobject.[hc]: implemented gtk_object_get() in terms of gtk_object_arg_get() and gtk_arg_to_valueloc(), floats are collected as gfloat*, uchars are collected as guchar*, ints are collected as gint*, etc... Mon Mar 29 17:45:47 1999 Owen Taylor <otaylor@redhat.com> * gtk/gtkaccelgroup.c gtk/gtkgtkbindings.c: Include <string.h> instead of <strings.h>. * gtk/gtkstyle.c: Fix double include of gtkthemes.h (actually, a lot more duplicate includes occur if you trace through the sequence of #include's) gtk-jbb-990320-0: John Bley, jbb6@acpub.duke.edu Mon Mar 29 17:02:58 1999 Owen Taylor <otaylor@redhat.com> Patches from Akira Higuchi <a-higuti@math.sci.hokudai.ac.jp> gtk-a-higuti-990322-[0-3] * configure.in: Fix confusion between GTK_LOCALE_[C]FLAGS that was causing -DX_LOCALE not to work. * gtk/gtkrc.c (gtk_rc_init): X_LOCALE will never have LC_MESSAGES defined Thu Mar 25 12:38:31 1999 Tim Janik <timj@gtk.org> * gtk/gtkrc.c (gtk_rc_append_default_module_path): get $HOME from g_get_home_dir() (gtk-pmc-990123-0.patch.gz). * gtk/gtkwindow.c (gtk_window_key_press_event): feature keypad up/down/ left/right as well (gtk-michael-980726-0.patch.gz). * gtk/gtklabel.[hc]: bunch of miscellaneous cleanups, such as s/0/NULL/ for pointer values, use gchar instead of char. fixed uline allocation leaks, changed the allocation pattern so we use G_ALLOC_AND_FREE mem chunks instead of G_ALLOC_ONLY. (gtk_label_size_request): always alter requisition as passed and leave widget->requisition alone. (gtk_label_set_text): allow NULL strings. (gtk_label_new): likewise. Wed Mar 24 09:24:03 1999 Tim Janik <timj@gtk.org> * gtk/gtkdrawingarea.[hc]: type/macro fixups. Mon Mar 22 05:51:34 1999 Tim Janik <timj@gtk.org> * gtk/gtkbin.c (gtk_bin_draw): only redraw children that are visible *and* mapped (i.e. drawable). (gtk_bin_expose): only send exposes to drawable children. * gtk/gtkbox.c (gtk_box_draw): only redraw children that are drawable. (gtk_box_expose): only send exposes to drawable children. * gtk/gtkhscale.c (gtk_hscale_draw): * gtk/gtkvscale.c (gtk_vscale_draw): hm, this is an ugly one. we first compute the size of our trough area here (window relative) and then check intersection with the draw_area which is parent relative because we're a NO_WINDOW widget, so we need to offset the trough area by allocation.x and allocation.y before the check. (this must not be done for the background area though, since that's already computed parent relative). Mon Mar 22 00:41:39 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_unrealize): unmap clist if neccessary, unrealize title buttons. Fri Mar 19 00:00:22 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.c (gtk_clist_column_title_passive) (gtk_clist_column_title_active): only connect/disconnect to GtkWidgetClass::event to block mouse events. * gtk/gtkclist.c (vertical_timeout) (horizontal_timeout): zero initialize event, removed superfluous gdk_window_get_pointer call * gtk/gtklist.c (gtk_list_vertical_timeout) (gtk_list_horizontal_timeout): removed superfluous gdk_window_get_pointer call Wed Mar 17 09:00:00 1999 Tim Janik <timj@gtk.org> * plugging problems reported by "Bruce Mitchener, Jr." <bruce@puremagic.com> due to a purify session. * gtk/gtkstyle.c: (gtk_style_ref): (gtk_style_unref): assert ref_count to be > 0. * gtk/gtkclist.c (gtk_clist_set_cell_style): { 0 } initilaize the requisition. (gtk_clist_set_shift): likewise. * gtk/gtklayout.c: introduce gtk_layout_finalize() to unref the adjustments. * gtk/gtklist.c (gtk_list_horizontal_timeout): zero initialize the event before sending it and set send_event to TRUE (which needs to be done for *all* synthesized events). (gtk_list_vertical_timeout): likewise. * gtk/gtktipsquery.c (gtk_tips_query_destroy): plug small memory leaks. * gtk/gtkdrawingarea.c (gtk_drawing_area_send_configure): set send_event to TRUE when synthesizing events. [ *** end of merges from 1.2 *** ]
1999-11-22 21:52:50 +00:00
g_assert (_gtk_private_n_signals < 65535);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal = LOOKUP_SIGNAL_ID (new_signal_id);
if (signal)
signal->signal_id = new_signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return signal;
}
static inline GtkHandler*
gtk_signal_get_handlers (GtkObject *object,
guint signal_id)
{
GtkHandler *handlers;
handlers = gtk_object_get_data_by_id (object, gtk_handler_quark);
while (handlers)
{
if (handlers->signal_id == signal_id)
return handlers;
handlers = handlers->next;
}
return NULL;
}
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
void
gtk_signal_init (void)
{
if (!gtk_handler_quark)
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
{
GtkSignal *zero;
zero = gtk_signal_next_and_invalidate ();
g_assert (zero == NULL);
gtk_handler_quark = g_quark_from_static_string ("gtk-signal-handlers");
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_hash_mem_chunk =
g_mem_chunk_new ("GtkSignalHash mem chunk",
sizeof (GtkSignalHash),
sizeof (GtkSignalHash) * SIGNAL_BLOCK_SIZE,
G_ALLOC_ONLY);
gtk_disconnect_info_mem_chunk =
g_mem_chunk_new ("GtkDisconnectInfo mem chunk",
sizeof (GtkDisconnectInfo),
sizeof (GtkDisconnectInfo) * DISCONNECT_INFO_BLOCK_SIZE,
G_ALLOC_AND_FREE);
gtk_handler_free_list = NULL;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
gtk_free_emissions = NULL;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_hash_table = g_hash_table_new (gtk_signal_hash,
gtk_signal_compare);
}
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint
gtk_signal_newv (const gchar *r_name,
new functions gtk_selection_data_copy and gtk_selection_data_free. Tue Jun 9 01:57:23 1998 Tim Janik <timj@gtk.org> * gtk/gtkselection.h: * gtk/gtkselection.c: new functions gtk_selection_data_copy and gtk_selection_data_free. * gtk/gtkwidget.c (gtk_widget_class_init): fixed gtk_signal_new() call for "selection_received", which was completely bogus. * other fixups to gtk_signal_new() calls all over the place. * gtk/gtktypebuiltins.h: types as variables (formerly macros). * gtk/gtktypebuiltins_vars.c: type variable implementations. * gtk/gtktypebuiltins_ids.c: array entries for builtin type declarations. * gtk/gtktypebuiltins_evals.c: enum value arrays. * gtk/gtk-boxed.defs: gtk and gdk structure definitions, used to build gtk.defs. * gtk/gtk.defs: generated file with scheme syntax for type definitions of gtk and gdk structures and enums. * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reworked type ids, so they are variables not macros anymore (this fixes binary incompatibility with new enum definitions). * gtk/gtkwidget.c (gtk_widget_real_key_press_event): proccess possible key bindings for this widget. * gtk/gtkwindow.c (gtk_window_key_press_event): chain parent class' handler. * gtk/gtkobject.h: * gtk/gtkobject.c: removed gtk_object_class_new_user_signal_no_recurse() again. new functions gtk_object_class_user_signal_new () and gtk_object_class_user_signal_newv (), to feature the GtkSignalRunType flag on the signal creation. Mon Jun 8 20:52:21 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: new signal GtkContainer::set_focus_child.
1998-06-09 07:11:55 +00:00
GtkSignalRunType signal_flags,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkType object_type,
guint function_offset,
GtkSignalMarshaller marshaller,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkType return_val,
guint nparams,
GtkType *params)
{
1997-11-24 22:37:52 +00:00
GtkSignal *signal;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkSignalHash *hash;
GQuark quark;
guint i;
gchar *name;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_val_if_fail (r_name != NULL, 0);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
g_return_val_if_fail (marshaller != NULL, 0);
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
g_return_val_if_fail (nparams < MAX_SIGNAL_PARAMS, 0);
if (nparams)
g_return_val_if_fail (params != NULL, 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
if (!gtk_handler_quark)
1997-11-24 22:37:52 +00:00
gtk_signal_init ();
name = g_strdup (r_name);
g_strdelimit (name, NULL, '_');
quark = gtk_signal_lookup (name, object_type);
if (quark)
1997-11-24 22:37:52 +00:00
{
1998-03-08 10:41:48 +00:00
g_warning ("gtk_signal_newv(): signal \"%s\" already exists in the `%s' class ancestry\n",
r_name,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_type_name (object_type));
g_free (name);
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
if (return_val != GTK_TYPE_NONE &&
new functions gtk_selection_data_copy and gtk_selection_data_free. Tue Jun 9 01:57:23 1998 Tim Janik <timj@gtk.org> * gtk/gtkselection.h: * gtk/gtkselection.c: new functions gtk_selection_data_copy and gtk_selection_data_free. * gtk/gtkwidget.c (gtk_widget_class_init): fixed gtk_signal_new() call for "selection_received", which was completely bogus. * other fixups to gtk_signal_new() calls all over the place. * gtk/gtktypebuiltins.h: types as variables (formerly macros). * gtk/gtktypebuiltins_vars.c: type variable implementations. * gtk/gtktypebuiltins_ids.c: array entries for builtin type declarations. * gtk/gtktypebuiltins_evals.c: enum value arrays. * gtk/gtk-boxed.defs: gtk and gdk structure definitions, used to build gtk.defs. * gtk/gtk.defs: generated file with scheme syntax for type definitions of gtk and gdk structures and enums. * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reworked type ids, so they are variables not macros anymore (this fixes binary incompatibility with new enum definitions). * gtk/gtkwidget.c (gtk_widget_real_key_press_event): proccess possible key bindings for this widget. * gtk/gtkwindow.c (gtk_window_key_press_event): chain parent class' handler. * gtk/gtkobject.h: * gtk/gtkobject.c: removed gtk_object_class_new_user_signal_no_recurse() again. new functions gtk_object_class_user_signal_new () and gtk_object_class_user_signal_newv (), to feature the GtkSignalRunType flag on the signal creation. Mon Jun 8 20:52:21 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: new signal GtkContainer::set_focus_child.
1998-06-09 07:11:55 +00:00
(signal_flags & GTK_RUN_BOTH) == GTK_RUN_FIRST)
{
Added a modular client-message-filter mechanism, that is used for the DND Sun Oct 18 18:16:39 1998 Owen Taylor <otaylor@gtk.org> * gdk/gdk.c gdkprivate.h: Added a modular client-message-filter mechanism, that is used for the DND messages. Removed all the old DND code. * gdk/gdkcolormap.c gdk/gdkcolormap.h: Add a function to get the visual of a given colormap. * gtk/gtkcolorsel.c: Conversion to new DND, drag a color-swatch. * gdk/gdk.h gdk/gdkdnd.c: The low-level X oriented portions of drag and drop protocols. Sending and receiving client messages, and navigating window trees. * gdk/gdkimage.c: added a gdk_flush() when destroying SHM images to hopefully make it more likely that X will gracefully handle the segment being destroyed. * gdk/gdkprivate.h gtk/gtkdebug.h: Add new DND debugging flags. * gtk/gtkeditable.[ch]: Updates for the selection handling changes. * gtk/gtkselection.[ch]: Added GtkTargetList, a refcounted data structure for keeping track of lists of GdkAtom + information. Removed selection_handler_add in favor of a "drag_data_get" signal. * gtk/gtkdnd.[ch] gtk/gtk.h: New files - highlevel (event loop dependent) parts of the DND protocols, display of drag icons, drag-under highlighting, and the "default handlers". * gtk/gtkinvisible.[ch]: New widget - InputOnly offscreen windows that are used for reliable pointer grabs and selection handling in the DND code. * gtk/testdnd.c: New test program for new DND. (Old DND tests in testgtk still need to be converted.) * gtk/testselection.c: Use the new selection API. * docs/dnd_internals: Start at describing how all the new code works inside. * docs/Changes-1.2.txt: New file describing source-incompatible changes in GTK+-1.2. Sat Oct 17 22:50:34 1998 Owen Taylor <otaylor@gtk.org> * gdk/gdkwindow.c (gdk_window_remove_filter): Free the right list node. * gdk/gdkwindow.c (gdk_window_init): Add gdk_root_parent to the XID table so we can receive events on it. Wed Oct 14 12:57:40 1998 Owen Taylor <otaylor@redhat.com> * gdk/gdk.c gdk/gdk.h (gdk_event_get_time): New function to get the timestamp from a generic event. Fri Oct 9 13:16:04 1998 Owen Taylor <otaylor@redhat.com> * gtk/gtkwidget.c (gtk_widget_add_events): Added function that safely adds additional events to a widget's event mask, even if the widget has previously been realized. (We can do this, but not remove events from the event mask). Fri Oct 2 17:35:35 1998 Owen Taylor <otaylor@redhat.com> * gdk/gdkproperty.c (gdk_property_get): Allow type == 0, for AnyPropertyType. Fri Oct 2 10:32:21 1998 Owen Taylor <otaylor@redhat.com> * gdk/gdkproperty.c (gdk_atom_intern): Add client-local hashing. Thu Sep 24 20:33:54 1998 Owen Taylor <otaylor@redhat.com> * gdk/gdk.c (gdk_event_send_clientmessage_toall): serial isn't a timestamp. Thu Sep 17 14:23:03 1998 Owen Taylor <otaylor@redhat.com> * gdk/gdk.c (gdk_event_translate): Removed printing of unknown window lookup warnings. (Made it a GDK_NOTE) - they happen in many circumstances.
1998-10-18 22:51:24 +00:00
g_warning ("gtk_signal_newv(): signal \"%s\" - return value `%s' incompatible with GTK_RUN_FIRST",
name, gtk_type_name (return_val));
g_free (name);
return 0;
}
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal = gtk_signal_next_and_invalidate ();
/* signal->signal_id already set */
signal->object_type = object_type;
signal->name = name;
1997-11-24 22:37:52 +00:00
signal->function_offset = function_offset;
signal->marshaller = marshaller;
signal->return_val = return_val;
signal->signal_flags = signal_flags;
1997-11-24 22:37:52 +00:00
signal->nparams = nparams;
signal->hook_list = NULL;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
if (nparams > 0)
{
signal->params = g_new (GtkType, nparams);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
for (i = 0; i < nparams; i++)
signal->params[i] = params[i];
1997-11-24 22:37:52 +00:00
}
else
signal->params = NULL;
/* insert "signal_name" into hash table
*/
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
hash = g_chunk_new (GtkSignalHash, gtk_signal_hash_mem_chunk);
hash->object_type = object_type;
hash->quark = g_quark_from_string (signal->name);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
hash->signal_id = signal->signal_id;
g_hash_table_insert (gtk_signal_hash_table, hash, GUINT_TO_POINTER (hash->signal_id));
/* insert "signal-name" into hash table
*/
g_strdelimit (signal->name, NULL, '-');
quark = g_quark_from_static_string (signal->name);
if (quark != hash->quark)
{
hash = g_chunk_new (GtkSignalHash, gtk_signal_hash_mem_chunk);
hash->object_type = object_type;
hash->quark = quark;
hash->signal_id = signal->signal_id;
g_hash_table_insert (gtk_signal_hash_table, hash, GUINT_TO_POINTER (hash->signal_id));
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return signal->signal_id;
}
guint
gtk_signal_new (const gchar *name,
new functions gtk_selection_data_copy and gtk_selection_data_free. Tue Jun 9 01:57:23 1998 Tim Janik <timj@gtk.org> * gtk/gtkselection.h: * gtk/gtkselection.c: new functions gtk_selection_data_copy and gtk_selection_data_free. * gtk/gtkwidget.c (gtk_widget_class_init): fixed gtk_signal_new() call for "selection_received", which was completely bogus. * other fixups to gtk_signal_new() calls all over the place. * gtk/gtktypebuiltins.h: types as variables (formerly macros). * gtk/gtktypebuiltins_vars.c: type variable implementations. * gtk/gtktypebuiltins_ids.c: array entries for builtin type declarations. * gtk/gtktypebuiltins_evals.c: enum value arrays. * gtk/gtk-boxed.defs: gtk and gdk structure definitions, used to build gtk.defs. * gtk/gtk.defs: generated file with scheme syntax for type definitions of gtk and gdk structures and enums. * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reworked type ids, so they are variables not macros anymore (this fixes binary incompatibility with new enum definitions). * gtk/gtkwidget.c (gtk_widget_real_key_press_event): proccess possible key bindings for this widget. * gtk/gtkwindow.c (gtk_window_key_press_event): chain parent class' handler. * gtk/gtkobject.h: * gtk/gtkobject.c: removed gtk_object_class_new_user_signal_no_recurse() again. new functions gtk_object_class_user_signal_new () and gtk_object_class_user_signal_newv (), to feature the GtkSignalRunType flag on the signal creation. Mon Jun 8 20:52:21 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: new signal GtkContainer::set_focus_child.
1998-06-09 07:11:55 +00:00
GtkSignalRunType signal_flags,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkType object_type,
guint function_offset,
GtkSignalMarshaller marshaller,
GtkType return_val,
guint nparams,
...)
{
GtkType *params;
guint i;
va_list args;
guint signal_id;
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
g_return_val_if_fail (nparams < MAX_SIGNAL_PARAMS, 0);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (nparams > 0)
{
params = g_new (GtkType, nparams);
va_start (args, nparams);
for (i = 0; i < nparams; i++)
params[i] = va_arg (args, GtkType);
va_end (args);
}
else
params = NULL;
signal_id = gtk_signal_newv (name,
new functions gtk_selection_data_copy and gtk_selection_data_free. Tue Jun 9 01:57:23 1998 Tim Janik <timj@gtk.org> * gtk/gtkselection.h: * gtk/gtkselection.c: new functions gtk_selection_data_copy and gtk_selection_data_free. * gtk/gtkwidget.c (gtk_widget_class_init): fixed gtk_signal_new() call for "selection_received", which was completely bogus. * other fixups to gtk_signal_new() calls all over the place. * gtk/gtktypebuiltins.h: types as variables (formerly macros). * gtk/gtktypebuiltins_vars.c: type variable implementations. * gtk/gtktypebuiltins_ids.c: array entries for builtin type declarations. * gtk/gtktypebuiltins_evals.c: enum value arrays. * gtk/gtk-boxed.defs: gtk and gdk structure definitions, used to build gtk.defs. * gtk/gtk.defs: generated file with scheme syntax for type definitions of gtk and gdk structures and enums. * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reworked type ids, so they are variables not macros anymore (this fixes binary incompatibility with new enum definitions). * gtk/gtkwidget.c (gtk_widget_real_key_press_event): proccess possible key bindings for this widget. * gtk/gtkwindow.c (gtk_window_key_press_event): chain parent class' handler. * gtk/gtkobject.h: * gtk/gtkobject.c: removed gtk_object_class_new_user_signal_no_recurse() again. new functions gtk_object_class_user_signal_new () and gtk_object_class_user_signal_newv (), to feature the GtkSignalRunType flag on the signal creation. Mon Jun 8 20:52:21 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: new signal GtkContainer::set_focus_child.
1998-06-09 07:11:55 +00:00
signal_flags,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
object_type,
function_offset,
marshaller,
return_val,
nparams,
params);
g_free (params);
return signal_id;
}
guint
gtk_signal_lookup (const gchar *name,
GtkType object_type)
{
GtkSignalHash hash;
GtkType lookup_type;
gpointer class = NULL;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
g_return_val_if_fail (name != NULL, 0);
g_return_val_if_fail (gtk_type_is_a (object_type, GTK_TYPE_OBJECT), 0);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
relookup:
lookup_type = object_type;
hash.quark = g_quark_try_string (name);
if (hash.quark)
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
{
while (lookup_type)
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
{
guint signal_id;
hash.object_type = lookup_type;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = GPOINTER_TO_UINT (g_hash_table_lookup (gtk_signal_hash_table, &hash));
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (signal_id)
return signal_id;
lookup_type = gtk_type_parent (lookup_type);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
}
}
if (!class)
{
class = gtk_type_class (object_type);
goto relookup;
}
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return 0;
1997-11-24 22:37:52 +00:00
}
GtkSignalQuery*
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
gtk_signal_query (guint signal_id)
{
GtkSignalQuery *query;
GtkSignal *signal;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_val_if_fail (signal_id >= 1, NULL);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal = LOOKUP_SIGNAL_ID (signal_id);
if (signal)
{
query = g_new (GtkSignalQuery, 1);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
query->object_type = signal->object_type;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
query->signal_id = signal_id;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
query->signal_name = signal->name;
query->is_user_signal = signal->function_offset == 0;
new functions gtk_selection_data_copy and gtk_selection_data_free. Tue Jun 9 01:57:23 1998 Tim Janik <timj@gtk.org> * gtk/gtkselection.h: * gtk/gtkselection.c: new functions gtk_selection_data_copy and gtk_selection_data_free. * gtk/gtkwidget.c (gtk_widget_class_init): fixed gtk_signal_new() call for "selection_received", which was completely bogus. * other fixups to gtk_signal_new() calls all over the place. * gtk/gtktypebuiltins.h: types as variables (formerly macros). * gtk/gtktypebuiltins_vars.c: type variable implementations. * gtk/gtktypebuiltins_ids.c: array entries for builtin type declarations. * gtk/gtktypebuiltins_evals.c: enum value arrays. * gtk/gtk-boxed.defs: gtk and gdk structure definitions, used to build gtk.defs. * gtk/gtk.defs: generated file with scheme syntax for type definitions of gtk and gdk structures and enums. * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reworked type ids, so they are variables not macros anymore (this fixes binary incompatibility with new enum definitions). * gtk/gtkwidget.c (gtk_widget_real_key_press_event): proccess possible key bindings for this widget. * gtk/gtkwindow.c (gtk_window_key_press_event): chain parent class' handler. * gtk/gtkobject.h: * gtk/gtkobject.c: removed gtk_object_class_new_user_signal_no_recurse() again. new functions gtk_object_class_user_signal_new () and gtk_object_class_user_signal_newv (), to feature the GtkSignalRunType flag on the signal creation. Mon Jun 8 20:52:21 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: new signal GtkContainer::set_focus_child.
1998-06-09 07:11:55 +00:00
query->signal_flags = signal->signal_flags;
query->return_val = signal->return_val;
query->nparams = signal->nparams;
query->params = signal->params;
}
else
query = NULL;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
return query;
}
1997-11-24 22:37:52 +00:00
gchar*
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
gtk_signal_name (guint signal_id)
1997-11-24 22:37:52 +00:00
{
GtkSignal *signal;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_val_if_fail (signal_id >= 1, NULL);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal = LOOKUP_SIGNAL_ID (signal_id);
1997-11-24 22:37:52 +00:00
if (signal)
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return signal->name;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
return NULL;
}
void
gtk_signal_emitv (GtkObject *object,
guint signal_id,
GtkArg *params)
{
GtkSignal *signal;
g_return_if_fail (object != NULL);
g_return_if_fail (signal_id >= 1);
signal = LOOKUP_SIGNAL_ID (signal_id);
g_return_if_fail (signal != NULL);
g_return_if_fail (gtk_type_is_a (GTK_OBJECT_TYPE (object), signal->object_type));
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
if (signal->nparams > 0)
g_return_if_fail (params != NULL);
gtk_signal_real_emit (object, signal_id, params);
}
void
1997-11-24 22:37:52 +00:00
gtk_signal_emit (GtkObject *object,
guint signal_id,
1997-11-24 22:37:52 +00:00
...)
{
GtkSignal *signal;
va_list args;
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
GtkArg params[MAX_SIGNAL_PARAMS + 1];
gboolean abort;
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (signal_id >= 1);
signal = LOOKUP_SIGNAL_ID (signal_id);
g_return_if_fail (signal != NULL);
g_return_if_fail (gtk_type_is_a (GTK_OBJECT_TYPE (object), signal->object_type));
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
va_start (args, signal_id);
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
abort = gtk_signal_collect_params (params,
signal->nparams,
signal->params,
signal->return_val,
args);
va_end (args);
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
if (!abort)
gtk_signal_real_emit (object, signal_id, params);
}
void
gtk_signal_emitv_by_name (GtkObject *object,
const gchar *name,
GtkArg *params)
{
guint signal_id;
g_return_if_fail (object != NULL);
g_return_if_fail (name != NULL);
g_return_if_fail (params != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
if (signal_id >= 1)
{
GtkSignal *signal;
signal = LOOKUP_SIGNAL_ID (signal_id);
g_return_if_fail (signal != NULL);
g_return_if_fail (gtk_type_is_a (GTK_OBJECT_TYPE (object), signal->object_type));
gtk_signal_real_emit (object, signal_id, params);
}
else
{
g_warning ("gtk_signal_emitv_by_name(): could not find signal \"%s\" in the `%s' class ancestry",
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
}
1997-11-24 22:37:52 +00:00
}
void
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_emit_by_name (GtkObject *object,
const gchar *name,
1997-11-24 22:37:52 +00:00
...)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (name != NULL);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (signal_id >= 1)
1997-11-24 22:37:52 +00:00
{
GtkSignal *signal;
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
GtkArg params[MAX_SIGNAL_PARAMS + 1];
va_list args;
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
gboolean abort;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
signal = LOOKUP_SIGNAL_ID (signal_id);
g_return_if_fail (signal != NULL);
g_return_if_fail (gtk_type_is_a (GTK_OBJECT_TYPE (object), signal->object_type));
va_start (args, name);
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
abort = gtk_signal_collect_params (params,
signal->nparams,
signal->params,
signal->return_val,
args);
1997-11-24 22:37:52 +00:00
va_end (args);
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
if (!abort)
gtk_signal_real_emit (object, signal_id, params);
1997-11-24 22:37:52 +00:00
}
1998-03-08 10:41:48 +00:00
else
{
g_warning ("gtk_signal_emit_by_name(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1998-03-08 10:41:48 +00:00
}
1997-11-24 22:37:52 +00:00
}
void
gtk_signal_emit_stop (GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
gint state;
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (signal_id >= 1);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
state = gtk_emission_check (current_emissions, object, signal_id);
if (state > 1)
g_warning ("gtk_signal_emit_stop(): emission (%u) for object `%s' cannot be stopped from emission hook",
signal_id,
gtk_type_name (GTK_OBJECT_TYPE (object)));
else if (state)
{
if (!gtk_emission_check (stop_emissions, object, signal_id))
gtk_emission_add (&stop_emissions, object, signal_id);
}
1998-03-08 10:41:48 +00:00
else
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_warning ("gtk_signal_emit_stop(): no current emission (%u) for object `%s'",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
}
void
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_emit_stop_by_name (GtkObject *object,
1997-11-24 22:37:52 +00:00
const gchar *name)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
g_return_if_fail (name != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (signal_id)
gtk_signal_emit_stop (object, signal_id);
1998-03-08 10:41:48 +00:00
else
g_warning ("gtk_signal_emit_stop_by_name(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
}
guint
gtk_signal_n_emissions (GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id)
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkEmission *emission;
guint n;
g_return_val_if_fail (object != NULL, 0);
g_return_val_if_fail (GTK_IS_OBJECT (object), 0);
n = 0;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
for (emission = current_emissions; emission; emission = emission->next)
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (emission->object == object && emission->signal_id == signal_id)
n++;
}
return n;
}
guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_n_emissions_by_name (GtkObject *object,
const gchar *name)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
guint n;
g_return_val_if_fail (object != NULL, 0);
g_return_val_if_fail (GTK_IS_OBJECT (object), 0);
g_return_val_if_fail (name != NULL, 0);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (signal_id)
n = gtk_signal_n_emissions (object, signal_id);
else
{
g_warning ("gtk_signal_n_emissions_by_name(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
n = 0;
}
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return n;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_connect (GtkObject *object,
const gchar *name,
1997-11-24 22:37:52 +00:00
GtkSignalFunc func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer func_data)
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_val_if_fail (object != NULL, 0);
g_return_val_if_fail (GTK_IS_OBJECT (object), 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (!signal_id)
1997-11-24 22:37:52 +00:00
{
1998-03-08 10:41:48 +00:00
g_warning ("gtk_signal_connect(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return gtk_signal_connect_by_type (object, signal_id,
1997-11-24 22:37:52 +00:00
func, func_data, NULL,
FALSE, FALSE, FALSE);
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_connect_after (GtkObject *object,
const gchar *name,
GtkSignalFunc func,
gpointer func_data)
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_val_if_fail (object != NULL, 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (!signal_id)
1997-11-24 22:37:52 +00:00
{
1998-03-08 10:41:48 +00:00
g_warning ("gtk_signal_connect_after(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return gtk_signal_connect_by_type (object, signal_id,
1997-11-24 22:37:52 +00:00
func, func_data, NULL,
FALSE, TRUE, FALSE);
1997-11-24 22:37:52 +00:00
}
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint
gtk_signal_connect_full (GtkObject *object,
const gchar *name,
GtkSignalFunc func,
GtkCallbackMarshal marshal,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer func_data,
GtkDestroyNotify destroy_func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gint object_signal,
gint after)
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_val_if_fail (object != NULL, 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (!signal_id)
1997-11-24 22:37:52 +00:00
{
1998-03-08 10:41:48 +00:00
g_warning ("gtk_signal_connect_full(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
if (marshal)
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return gtk_signal_connect_by_type (object, signal_id, (GtkSignalFunc) marshal,
func_data, destroy_func,
object_signal, after, TRUE);
else
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return gtk_signal_connect_by_type (object, signal_id, func,
func_data, destroy_func,
object_signal, after, FALSE);
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_connect_object (GtkObject *object,
const gchar *name,
1997-11-24 22:37:52 +00:00
GtkSignalFunc func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkObject *slot_object)
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_val_if_fail (object != NULL, 0);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
/* slot_object needs to be treated as ordinary pointer
*/
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (!signal_id)
1997-11-24 22:37:52 +00:00
{
1998-03-08 10:41:48 +00:00
g_warning ("gtk_signal_connect_object(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return gtk_signal_connect_by_type (object, signal_id,
1997-11-24 22:37:52 +00:00
func, slot_object, NULL,
TRUE, FALSE, FALSE);
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint
1997-11-24 22:37:52 +00:00
gtk_signal_connect_object_after (GtkObject *object,
const gchar *name,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkSignalFunc func,
1997-11-24 22:37:52 +00:00
GtkObject *slot_object)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint signal_id;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_val_if_fail (object != NULL, 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id = gtk_signal_lookup (name, GTK_OBJECT_TYPE (object));
if (!signal_id)
1997-11-24 22:37:52 +00:00
{
1998-03-08 10:41:48 +00:00
g_warning ("gtk_signal_connect_object_after(): could not find signal \"%s\" in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
name,
gtk_type_name (GTK_OBJECT_TYPE (object)));
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return gtk_signal_connect_by_type (object, signal_id,
1997-11-24 22:37:52 +00:00
func, slot_object, NULL,
TRUE, TRUE, FALSE);
1997-11-24 22:37:52 +00:00
}
void
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_connect_while_alive (GtkObject *object,
const gchar *signal,
GtkSignalFunc func,
gpointer func_data,
GtkObject *alive_object)
{
GtkDisconnectInfo *info;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (object != NULL);
g_return_if_fail (GTK_IS_OBJECT (object));
g_return_if_fail (signal != NULL);
g_return_if_fail (func != NULL);
g_return_if_fail (alive_object != NULL);
g_return_if_fail (GTK_IS_OBJECT (alive_object));
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
info = g_chunk_new (GtkDisconnectInfo, gtk_disconnect_info_mem_chunk);
info->object1 = object;
info->object2 = alive_object;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
info->signal_handler = gtk_signal_connect (object, signal, func, func_data);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
info->disconnect_handler1 =
gtk_signal_connect_object (info->object1,
"destroy",
GTK_SIGNAL_FUNC (gtk_alive_disconnecter),
(GtkObject*) info);
info->disconnect_handler2 =
gtk_signal_connect_object (info->object2,
"destroy",
GTK_SIGNAL_FUNC (gtk_alive_disconnecter),
(GtkObject*) info);
}
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
void
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_connect_object_while_alive (GtkObject *object,
const gchar *signal,
GtkSignalFunc func,
GtkObject *alive_object)
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
{
GtkDisconnectInfo *info;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
g_return_if_fail (object != NULL);
g_return_if_fail (GTK_IS_OBJECT (object));
g_return_if_fail (signal != NULL);
g_return_if_fail (func != NULL);
g_return_if_fail (alive_object != NULL);
g_return_if_fail (GTK_IS_OBJECT (alive_object));
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
info = g_chunk_new (GtkDisconnectInfo, gtk_disconnect_info_mem_chunk);
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
info->object1 = object;
info->object2 = alive_object;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
info->signal_handler = gtk_signal_connect_object (object, signal, func, alive_object);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
info->disconnect_handler1 =
gtk_signal_connect_object (info->object1,
"destroy",
GTK_SIGNAL_FUNC (gtk_alive_disconnecter),
(GtkObject*) info);
info->disconnect_handler2 =
gtk_signal_connect_object (info->object2,
"destroy",
GTK_SIGNAL_FUNC (gtk_alive_disconnecter),
(GtkObject*) info);
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
}
1997-11-24 22:37:52 +00:00
void
gtk_signal_disconnect (GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
guint handler_id)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handler;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (handler_id > 0);
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
while (handler)
1997-11-24 22:37:52 +00:00
{
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
if (handler->id == handler_id)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler->id = 0;
handler->blocked += 1;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
gtk_signal_handler_unref (handler, object);
1997-11-24 22:37:52 +00:00
return;
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = handler->next;
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_warning ("gtk_signal_disconnect(): could not find handler (%u)", handler_id);
1997-11-24 22:37:52 +00:00
}
void
gtk_signal_disconnect_by_func (GtkObject *object,
GtkSignalFunc func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer data)
{
GtkHandler *handler;
gint found_one;
g_return_if_fail (object != NULL);
g_return_if_fail (func != NULL);
found_one = FALSE;
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
while (handler)
{
GtkHandler *handler_next;
handler_next = handler->next;
if ((handler->id > 0) &&
(handler->func == func) &&
(handler->func_data == data))
{
found_one = TRUE;
handler->id = 0;
handler->blocked += 1;
gtk_signal_handler_unref (handler, object);
}
handler = handler_next;
}
if (!found_one)
g_warning ("gtk_signal_disconnect_by_func(): could not find handler (0x%0lX) containing data (0x%0lX)", (long) func, (long) data);
}
1997-11-24 22:37:52 +00:00
void
gtk_signal_disconnect_by_data (GtkObject *object,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer data)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handler;
1997-11-24 22:37:52 +00:00
gint found_one;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
found_one = FALSE;
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
while (handler)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handler_next;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler_next = handler->next;
if ((handler->id > 0) &&
(handler->func_data == data))
1997-11-24 22:37:52 +00:00
{
found_one = TRUE;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler->id = 0;
handler->blocked += 1;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
gtk_signal_handler_unref (handler, object);
1997-11-24 22:37:52 +00:00
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = handler_next;
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
if (!found_one)
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
g_warning ("gtk_signal_disconnect_by_data(): could not find handler containing data (0x%0lX)", (long) data);
1997-11-24 22:37:52 +00:00
}
void
gtk_signal_handler_block (GtkObject *object,
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint handler_id)
1997-11-24 22:37:52 +00:00
{
GtkHandler *handler;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (handler_id > 0);
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
while (handler)
1997-11-24 22:37:52 +00:00
{
if (handler->id == handler_id)
1997-11-24 22:37:52 +00:00
{
handler->blocked += 1;
1997-11-24 22:37:52 +00:00
return;
}
handler = handler->next;
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_warning ("gtk_signal_handler_block(): could not find handler (%u)", handler_id);
1997-11-24 22:37:52 +00:00
}
void
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_handler_block_by_func (GtkObject *object,
GtkSignalFunc func,
gpointer data)
{
GtkHandler *handler;
gint found_one;
g_return_if_fail (object != NULL);
g_return_if_fail (func != NULL);
found_one = FALSE;
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
while (handler)
{
if ((handler->id > 0) &&
(handler->func == func) &&
(handler->func_data == data))
{
found_one = TRUE;
handler->blocked += 1;
}
handler = handler->next;
}
if (!found_one)
g_warning ("gtk_signal_handler_block_by_func(): could not find handler (0x%0lX) containing data (0x%0lX)", (long) func, (long) data);
}
1997-11-24 22:37:52 +00:00
void
gtk_signal_handler_block_by_data (GtkObject *object,
gpointer data)
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handler;
1997-11-24 22:37:52 +00:00
gint found_one;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
found_one = FALSE;
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
while (handler)
1997-11-24 22:37:52 +00:00
{
if ((handler->id > 0) &&
(handler->func_data == data))
1997-11-24 22:37:52 +00:00
{
found_one = TRUE;
handler->blocked += 1;
1997-11-24 22:37:52 +00:00
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = handler->next;
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
if (!found_one)
GTK_RESIZE_NEEDED is a private flag now. Mon Feb 2 04:15:08 1998 Tim Janik <timj@gimp.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: GTK_RESIZE_NEEDED is a private flag now. (gtk_container_register_toplevel): new function. (gtk_container_unregister_toplevel): new function. * gtk/gtkmain.c: GTK_LEAVE_PENDING is a private flag now. * gtk/gtkmenu.c: call gtk_container_register_toplevel in gtk_menu_class_init instead of this dirty gtk_widget_set_parent(,NULL) hack. new default handler gtk_menu_destroy for calling gtk_container_unregister_toplevel. removed GTK_ANCHORED, GTK_UNMAPPED. * gtk/gtkobject.h: macro cleanups, added GTK_DESTROYED flag. * gtk/gtkobject.c: only emit DESTROY signal if !GTK_OBJECT_DESTROYED (object). * gtk/gtkprivate.h: new file that will not be automatically included. it holds the private flags for GtkWidget along with it's SET/UNSET and examination macros. * gtk/gtkwidget.c: private flags: GTK_RESIZE_NEEDED, GTK_REDRAW_PENDING, GTK_RESIZE_PENDING, GTK_IN_REPARENT, GTK_USER_STYLE. GTK_ANCHORED is replaced by GTK_TOPLEVEL. added missing UNSET for GTK_IN_REPARENT. removed the gtk_widget_set_parent(, NULL) hack for toplevels. upon destroy free memory for widgets with GTK_WIDGET_HAS_SHAPE_MASK. * gtk/gtkwidget.h: split up the widget flags into a public and a private portion. added an extra field private_flags to GtkWidget without making it bigger by using an alignment gap of 16 bit. macro cleanups. * gtk/gtkwindow.c: removed GTK_ANCHORED. new function gtk_window_destroy for calling gtk_container_unregister_toplevel. removed the gtk_widget_set_parent(,NULL), call gtk_container_register_toplevel instead. remove GTK_UNMAPPED. GTK_RESIZE_NEEDED is private now. * gtk/gtksignal.c (gtk_signal_disconnect): removed a bug on removal that cut off the handler list -> living_objects == 0 with testgtk. made some warnings more descriptive. new function gtk_signal_connect_object_while_alive, which will automatically destroy the connection once one of the objects is destroyed. didn't include this before removal of the above mentioned bug. * reflected refcounting revolution in ChangeLog
1998-02-02 04:54:25 +00:00
g_warning ("gtk_signal_handler_block_by_data(): could not find handler containing data (0x%0lX)", (long) data);
1997-11-24 22:37:52 +00:00
}
void
gtk_signal_handler_unblock (GtkObject *object,
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint handler_id)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handler;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_if_fail (handler_id > 0);
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
while (handler)
1997-11-24 22:37:52 +00:00
{
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
if (handler->id == handler_id)
1997-11-24 22:37:52 +00:00
{
if (handler->blocked > 0)
handler->blocked -= 1;
else
g_warning ("gtk_signal_handler_unblock(): handler (%u) is not blocked", handler_id);
1997-11-24 22:37:52 +00:00
return;
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = handler->next;
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_warning ("gtk_signal_handler_unblock(): could not find handler (%u)", handler_id);
1997-11-24 22:37:52 +00:00
}
void
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_handler_unblock_by_func (GtkObject *object,
GtkSignalFunc func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer data)
{
GtkHandler *handler;
gint found_one;
g_return_if_fail (object != NULL);
g_return_if_fail (func != NULL);
found_one = FALSE;
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
while (handler)
{
if ((handler->id > 0) &&
(handler->func == func) &&
(handler->func_data == data) &&
(handler->blocked > 0))
{
handler->blocked -= 1;
found_one = TRUE;
}
handler = handler->next;
}
if (!found_one)
g_warning ("gtk_signal_handler_unblock_by_func(): could not find blocked handler (0x%0lX) containing data (0x%0lX)", (long) func, (long) data);
}
1997-11-24 22:37:52 +00:00
void
gtk_signal_handler_unblock_by_data (GtkObject *object,
gpointer data)
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handler;
1997-11-24 22:37:52 +00:00
gint found_one;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
found_one = FALSE;
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
while (handler)
1997-11-24 22:37:52 +00:00
{
if ((handler->id > 0) &&
(handler->func_data == data) &&
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
(handler->blocked > 0))
1997-11-24 22:37:52 +00:00
{
handler->blocked -= 1;
1997-11-24 22:37:52 +00:00
found_one = TRUE;
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = handler->next;
1997-11-24 22:37:52 +00:00
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
if (!found_one)
g_warning ("gtk_signal_handler_unblock_by_data(): could not find blocked handler containing data (0x%0lX)", (long) data);
1997-11-24 22:37:52 +00:00
}
void
gtk_signal_handlers_destroy (GtkObject *object)
{
GtkHandler *handler;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
/* we make the "optimization" of destroying the first handler in the last
* place, since we don't want gtk_signal_handler_unref() to reset the objects
* handler_key data on each removal
*/
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
if (handler)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = handler->next;
while (handler)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *next;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
next = handler->next;
if (handler->id > 0)
gtk_signal_handler_unref (handler, object);
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler = next;
1997-11-24 22:37:52 +00:00
}
handler = gtk_object_get_data_by_id (object, gtk_handler_quark);
if (handler->id > 0)
gtk_signal_handler_unref (handler, object);
1997-11-24 22:37:52 +00:00
}
}
void
gtk_signal_set_funcs (GtkSignalMarshal marshal_func,
GtkSignalDestroy destroy_func)
{
global_marshaller = marshal_func;
global_destroy_notify = destroy_func;
1997-11-24 22:37:52 +00:00
}
static guint
gtk_signal_hash (gconstpointer h)
1997-11-24 22:37:52 +00:00
{
Removed g_object_pointer_hash, which was just g_direct_hash. Tue Jun 9 18:44:57 1998 Owen Taylor <otaylor@gtk.org> * gtk/gtkobject.c: Removed g_object_pointer_hash, which was just g_direct_hash. Mon May 25 19:54:20 1998 Owen Taylor <otaylor@gtk.org> * configure.in: x_libs=, not $x_libs=. Enough said. (Case only hit for --disable-xshm) Mon May 25 12:08:14 1998 Owen Taylor <otaylor@gtk.org> * configure.in (LDFLAGS): Add to $CFLAGS and $LDFLAGS when testing for X libraries, don't replace them. Because the user might have specified the path to the X libraries themself before running configure. * examples/**.c: Changed all gpointer * to gpointer Sat May 23 21:54:05 1998 Owen Taylor <otaylor@gtk.org> * configure.in (LDFLAGS): Bomb out with a moderately helpful message if detection of X libraries fails. Sat May 23 18:57:06 1998 Owen Taylor <otaylor@gtk.org> [ Combination of: gtk-rrh-980412-0.patch (Raja R Harinath <harinath@cs.umn.edu>) gtk-jbuhler-980516-0 (Jeremy Buhler <jbuhler@cs.washington.edu>) ] * gdk/gdk.h gdk/gdkcc.c gdk/gdkfont.c gtk/gtkmain.c gtk/gtksignal.c gtk/gtktext.c: Fixups for warnings from adding const to type of GHashFunc, GCompareFunc * gtk/gtkcombo.c (gtk_combo_entry_key_press): Minor style/ ansi-warnings fixups. Tue Jun 9 17:47:33 1998 Owen Taylor <otaylor@gtk.org> * glib.h: Remove #error - HP/UX. Sat May 23 19:00:01 1998 Owen Taylor <otaylor@gtk.org> [ Combination of: gtk-rrh-980412-0.patch (Raja R Harinath <harinath@cs.umn.edu>) gtk-jbuhler-980516-0 (Jeremy Buhler <jbuhler@cs.washington.edu>) ] * glib.h ghash.c gstring.c gdataset.c gutils.c: - Added new typedef g_const_pointer; expunged all incorrect uses of 'const gpointer'. - Fixed up warnings that that created, - Changed GHashFunc and GCompareFunc to take g_const_pointer arguments. (Necessary, but will cause warnings in existing code until fixed) - Added other new const in harmless positions.
1998-06-09 23:18:11 +00:00
register const GtkSignalHash *hash = h;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return hash->object_type ^ hash->quark;
1997-11-24 22:37:52 +00:00
}
static gint
gtk_signal_compare (gconstpointer h1,
gconstpointer h2)
1997-11-24 22:37:52 +00:00
{
Removed g_object_pointer_hash, which was just g_direct_hash. Tue Jun 9 18:44:57 1998 Owen Taylor <otaylor@gtk.org> * gtk/gtkobject.c: Removed g_object_pointer_hash, which was just g_direct_hash. Mon May 25 19:54:20 1998 Owen Taylor <otaylor@gtk.org> * configure.in: x_libs=, not $x_libs=. Enough said. (Case only hit for --disable-xshm) Mon May 25 12:08:14 1998 Owen Taylor <otaylor@gtk.org> * configure.in (LDFLAGS): Add to $CFLAGS and $LDFLAGS when testing for X libraries, don't replace them. Because the user might have specified the path to the X libraries themself before running configure. * examples/**.c: Changed all gpointer * to gpointer Sat May 23 21:54:05 1998 Owen Taylor <otaylor@gtk.org> * configure.in (LDFLAGS): Bomb out with a moderately helpful message if detection of X libraries fails. Sat May 23 18:57:06 1998 Owen Taylor <otaylor@gtk.org> [ Combination of: gtk-rrh-980412-0.patch (Raja R Harinath <harinath@cs.umn.edu>) gtk-jbuhler-980516-0 (Jeremy Buhler <jbuhler@cs.washington.edu>) ] * gdk/gdk.h gdk/gdkcc.c gdk/gdkfont.c gtk/gtkmain.c gtk/gtksignal.c gtk/gtktext.c: Fixups for warnings from adding const to type of GHashFunc, GCompareFunc * gtk/gtkcombo.c (gtk_combo_entry_key_press): Minor style/ ansi-warnings fixups. Tue Jun 9 17:47:33 1998 Owen Taylor <otaylor@gtk.org> * glib.h: Remove #error - HP/UX. Sat May 23 19:00:01 1998 Owen Taylor <otaylor@gtk.org> [ Combination of: gtk-rrh-980412-0.patch (Raja R Harinath <harinath@cs.umn.edu>) gtk-jbuhler-980516-0 (Jeremy Buhler <jbuhler@cs.washington.edu>) ] * glib.h ghash.c gstring.c gdataset.c gutils.c: - Added new typedef g_const_pointer; expunged all incorrect uses of 'const gpointer'. - Fixed up warnings that that created, - Changed GHashFunc and GCompareFunc to take g_const_pointer arguments. (Necessary, but will cause warnings in existing code until fixed) - Added other new const in harmless positions.
1998-06-09 23:18:11 +00:00
register const GtkSignalHash *hash1 = h1;
register const GtkSignalHash *hash2 = h2;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return (hash1->quark == hash2->quark &&
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
hash1->object_type == hash2->object_type);
1997-11-24 22:37:52 +00:00
}
static guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_alive_disconnecter (GtkDisconnectInfo *info)
1997-11-24 22:37:52 +00:00
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
g_return_val_if_fail (info != NULL, 0);
gtk_signal_disconnect (info->object1, info->disconnect_handler1);
gtk_signal_disconnect (info->object1, info->signal_handler);
gtk_signal_disconnect (info->object2, info->disconnect_handler2);
g_mem_chunk_free (gtk_disconnect_info_mem_chunk, info);
return 0;
1997-11-24 22:37:52 +00:00
}
static GtkHandler*
configure.in acheader.h gdk/gdkwindow.c Check for Shape extension both on Sun May 3 13:38:22 1998 Owen Taylor <otaylor@gtk.org> * configure.in acheader.h gdk/gdkwindow.c Check for Shape extension both on the client and server side. (And, more importantly, check for the shape extension so we may include -lXext even when compiling with --disable-xshm) Don't set override_redirect on all shaped windows. It isn't necessary. * gdk/gdkwindow.c: Set ->colormap to NULL for root and foreign windows. Use this to check if we need to get the colormap from X. Fri May 1 22:32:47 1998 Owen Taylor <otaylor@gtk.org> * gtk/gtkbutton.c (gtk_button_paint): Draw the areas between the default and the button always in GTK_STATE_NORMAL. * gtk/gtkrange.c (gtk_range_style_set): Added a style_set callback. Fri May 1 16:40:57 1998 Owen Taylor <otaylor@gtk.org> * gdk/gdkpixmap.c (gdk_pixmap_colormap_create_from_xpmp[_d]): Fix a buffer overflow on pixmaps that claim to have more than 31 characters per pixel. (gdk_pixmap_read_string): Don't wrap around strings longer than half of address space ;-) * gtk/gtk[vh]ruler.c gtk/gtkinputdialog.c: Expand some buffers that were used for printing integers. * */* (almost): Style: All int foo () { ... } changed to int foo (void) { ... } ^^^^^^^ This is why some many files changed Even where there were proper prototypes elsewhere. * gdk/gxid.c (handle_claim_device): Some extra checks. It isn't safe against being fed bad X id's, but at least it should be safe against deleting all your files.
1998-05-03 22:41:32 +00:00
gtk_signal_handler_new (void)
1997-11-24 22:37:52 +00:00
{
GtkHandler *handler;
if (!gtk_handler_free_list)
{
GtkHandler *handler_block;
guint i;
handler_block = g_new0 (GtkHandler, HANDLER_BLOCK_SIZE);
for (i = 1; i < HANDLER_BLOCK_SIZE; i++)
{
(handler_block + i)->next = gtk_handler_free_list;
gtk_handler_free_list = (handler_block + i);
}
handler = handler_block;
}
else
{
handler = gtk_handler_free_list;
gtk_handler_free_list = handler->next;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
handler->id = 0;
handler->blocked = 0;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
handler->signal_id = 0;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler->object_signal = FALSE;
1997-11-24 22:37:52 +00:00
handler->after = FALSE;
handler->no_marshal = FALSE;
handler->ref_count = 1;
1997-11-24 22:37:52 +00:00
handler->func = NULL;
handler->func_data = NULL;
handler->destroy_func = NULL;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler->prev = NULL;
1997-11-24 22:37:52 +00:00
handler->next = NULL;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
return handler;
}
static void
gtk_signal_handler_ref (GtkHandler *handler)
1997-11-24 22:37:52 +00:00
{
handler->ref_count += 1;
}
static void
gtk_signal_handler_unref (GtkHandler *handler,
GtkObject *object)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (!handler->ref_count)
{
/* FIXME: i wanna get removed somewhen */
g_warning ("gtk_signal_handler_unref(): handler with ref_count==0!");
return;
}
handler->ref_count -= 1;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (handler->ref_count == 0)
{
if (handler->destroy_func)
(* handler->destroy_func) (handler->func_data);
else if (!handler->func && global_destroy_notify)
(* global_destroy_notify) (handler->func_data);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
if (handler->prev)
handler->prev->next = handler->next;
else if (handler->next)
gtk_object_set_data_by_id (object, gtk_handler_quark, handler->next);
else
{
GTK_OBJECT_UNSET_FLAGS (object, GTK_CONNECTED);
gtk_object_set_data_by_id (object, gtk_handler_quark, NULL);
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
if (handler->next)
handler->next->prev = handler->prev;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
handler->next = gtk_handler_free_list;
gtk_handler_free_list = handler;
}
1997-11-24 22:37:52 +00:00
}
static void
gtk_signal_handler_insert (GtkObject *object,
GtkHandler *handler)
{
GtkHandler *tmp;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
/* FIXME: remove */ g_assert (handler->next == NULL);
/* FIXME: remove */ g_assert (handler->prev == NULL);
tmp = gtk_object_get_data_by_id (object, gtk_handler_quark);
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
if (!tmp)
{
GTK_OBJECT_SET_FLAGS (object, GTK_CONNECTED);
gtk_object_set_data_by_id (object, gtk_handler_quark, handler);
}
1997-11-24 22:37:52 +00:00
else
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
while (tmp)
{
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (tmp->signal_id < handler->signal_id)
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
{
if (tmp->prev)
{
tmp->prev->next = handler;
handler->prev = tmp->prev;
}
else
gtk_object_set_data_by_id (object, gtk_handler_quark, handler);
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
tmp->prev = handler;
handler->next = tmp;
break;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
if (!tmp->next)
{
tmp->next = handler;
handler->prev = tmp;
break;
}
tmp = tmp->next;
}
1997-11-24 22:37:52 +00:00
}
#ifdef G_ENABLE_DEBUG
/* value typically set via gdb */
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
static GtkObject *gtk_trace_signal_object = NULL;
#endif /* G_ENABLE_DEBUG */
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
static void
1997-11-24 22:37:52 +00:00
gtk_signal_real_emit (GtkObject *object,
guint signal_id,
GtkArg *params)
1997-11-24 22:37:52 +00:00
{
GtkSignal signal;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
GtkHandler *handlers;
GtkSignalFunc signal_func;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkEmission *emission;
/* gtk_handlers_run() expects a reentrant GtkSignal*, so we allocate
* it locally on the stack. we save some lookups ourselves with this as well.
*/
signal = *LOOKUP_SIGNAL_ID (signal_id);
if (signal.function_offset)
signal_func = G_STRUCT_MEMBER (GtkSignalFunc, object->klass, signal.function_offset);
else
signal_func = NULL;
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
#ifdef G_ENABLE_DEBUG
if (gtk_debug_flags & GTK_DEBUG_SIGNALS ||
object == gtk_trace_signal_object)
g_message ("%s::%s emitted (object=%p class-method=%p)\n",
gtk_type_name (GTK_OBJECT_TYPE (object)),
signal.name,
object,
signal_func);
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
#endif /* G_ENABLE_DEBUG */
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (signal.signal_flags & GTK_RUN_NO_RECURSE)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
gint state;
state = gtk_emission_check (current_emissions, object, signal_id);
if (state)
{
if (state > 1)
g_warning ("gtk_signal_real_emit(): emission (%u) for object `%s' cannot be restarted from emission hook",
signal_id,
gtk_type_name (GTK_OBJECT_TYPE (object)));
else if (!gtk_emission_check (restart_emissions, object, signal_id))
gtk_emission_add (&restart_emissions, object, signal_id);
return;
}
1997-11-24 22:37:52 +00:00
}
gtk_object_ref (object);
gtk_emission_add (&current_emissions, object, signal_id);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission = current_emissions;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
emission_restart:
if (signal.signal_flags & GTK_RUN_FIRST && signal_func)
{
signal.marshaller (object, signal_func, NULL, params);
if (stop_emissions && gtk_emission_check (stop_emissions, object, signal_id))
{
gtk_emission_remove (&stop_emissions, object, signal_id);
goto emission_done;
}
else if (restart_emissions &&
signal.signal_flags & GTK_RUN_NO_RECURSE &&
gtk_emission_check (restart_emissions, object, signal_id))
{
gtk_emission_remove (&restart_emissions, object, signal_id);
goto emission_restart;
}
}
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (signal.hook_list && !GTK_OBJECT_DESTROYED (object))
{
GtkEmissionHookData data;
data.object = object;
data.n_params = signal.nparams;
data.params = params;
data.signal_id = signal_id;
emission->in_hook = 1;
g_hook_list_marshal_check (signal.hook_list, TRUE, gtk_emission_hook_marshaller, &data);
emission->in_hook = 0;
}
if (GTK_OBJECT_CONNECTED (object))
{
handlers = gtk_signal_get_handlers (object, signal_id);
if (handlers)
{
gint return_val;
return_val = gtk_handlers_run (handlers, &signal, object, params, FALSE);
switch (return_val)
{
case EMISSION_CONTINUE:
break;
case EMISSION_RESTART:
goto emission_restart;
case EMISSION_DONE:
goto emission_done;
}
}
}
if (signal.signal_flags & GTK_RUN_LAST && signal_func)
{
signal.marshaller (object, signal_func, NULL, params);
if (stop_emissions && gtk_emission_check (stop_emissions, object, signal_id))
{
gtk_emission_remove (&stop_emissions, object, signal_id);
goto emission_done;
}
else if (restart_emissions &&
signal.signal_flags & GTK_RUN_NO_RECURSE &&
gtk_emission_check (restart_emissions, object, signal_id))
{
gtk_emission_remove (&restart_emissions, object, signal_id);
goto emission_restart;
}
}
if (GTK_OBJECT_CONNECTED (object))
{
handlers = gtk_signal_get_handlers (object, signal_id);
if (handlers)
{
gint return_val;
return_val = gtk_handlers_run (handlers, &signal, object, params, TRUE);
switch (return_val)
{
case EMISSION_CONTINUE:
break;
case EMISSION_RESTART:
goto emission_restart;
case EMISSION_DONE:
goto emission_done;
}
}
}
emission_done:
if (restart_emissions && signal.signal_flags & GTK_RUN_NO_RECURSE)
gtk_emission_remove (&restart_emissions, object, signal_id);
gtk_emission_remove (&current_emissions, object, signal_id);
gtk_object_unref (object);
1997-11-24 22:37:52 +00:00
}
guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_handler_pending (GtkObject *object,
guint signal_id,
gboolean may_be_blocked)
{
GtkHandler *handlers;
guint handler_id;
g_return_val_if_fail (object != NULL, 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_return_val_if_fail (signal_id >= 1, 0);
if (GTK_OBJECT_CONNECTED (object))
handlers = gtk_signal_get_handlers (object, signal_id);
else
return 0;
handler_id = 0;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
while (handlers && handlers->signal_id == signal_id)
{
if (handlers->id > 0 &&
yeppers, accelerator changes to fix the gimp. commit message dedicated to yeppers, accelerator changes to fix the gimp. commit message dedicated to sopwith ;) Thu Jun 18 03:30:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new function gtk_accel_label_accelerator_width to request the size of the accelerator portion of an accel label. (gtk_accel_label_size_request): don't request for the accelerators size. (gtk_accel_label_expose_event): only draw the accelerator if we got enough extra space. * gtk/gtkmenuitem.c (gtk_menu_item_size_request): request accelerator width from children. * gtk/gtkmenu.c (gtk_menu_key_press): when adding an accelerator to an object (after removal has been requested) check if there is still an accelerator remaining to avoid adding two accelerators on an object. this can happen for locked accelerators (or accelerator-frozen widgets). (gtk_menu_size_request): feature childrens accelerator width in size requests. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): use gtk_widget_freeze_accelerators() for dynamically created menu items. * gtk/gtksignal.h: * gtk/gtksignal.c: new function gtk_signal_handler_pending_by_func() which will return a handler_id > 0 if the specified function is pending for `signal_id'. * gtk/gtkwidget.h: * gtk/gtkwidget.c: remove gtk_widget_stop_accelerator, which was just a signal handler function to stop accelerator addition. added gtk_widget_freeze_accelerators and gtk_widget_thaw_accelerators which will prevent (undo) any accelerators from being added to or removed from a widget.
1998-06-18 03:22:09 +00:00
(may_be_blocked || handlers->blocked == FALSE))
{
handler_id = handlers->id;
break;
}
handlers = handlers->next;
}
return handler_id;
}
guint
gtk_signal_handler_pending_by_func (GtkObject *object,
guint signal_id,
gboolean may_be_blocked,
GtkSignalFunc func,
gpointer data)
{
GtkHandler *handlers;
guint handler_id;
g_return_val_if_fail (object != NULL, 0);
g_return_val_if_fail (func != NULL, 0);
g_return_val_if_fail (signal_id >= 1, 0);
if (GTK_OBJECT_CONNECTED (object))
handlers = gtk_signal_get_handlers (object, signal_id);
else
return 0;
handler_id = 0;
while (handlers && handlers->signal_id == signal_id)
{
if (handlers->id > 0 &&
handlers->func == func &&
handlers->func_data == data &&
(may_be_blocked || handlers->blocked == 0))
{
handler_id = handlers->id;
break;
}
handlers = handlers->next;
}
return handler_id;
}
gint
gtk_signal_handler_pending_by_id (GtkObject *object,
guint handler_id,
gboolean may_be_blocked)
{
GtkHandler *handlers;
g_return_val_if_fail (object != NULL, FALSE);
g_return_val_if_fail (handler_id >= 1, FALSE);
if (GTK_OBJECT_CONNECTED (object))
handlers = gtk_object_get_data_by_id (object, gtk_handler_quark);
else
return FALSE;
while (handlers)
{
if (handlers->id == handler_id)
return may_be_blocked || handlers->blocked == 0;
handlers = handlers->next;
}
return FALSE;
}
guint
gtk_signal_add_emission_hook (guint signal_id,
GtkEmissionHook hook_func,
gpointer data)
{
return gtk_signal_add_emission_hook_full (signal_id, hook_func, data, NULL);
}
guint
gtk_signal_add_emission_hook_full (guint signal_id,
GtkEmissionHook hook_func,
gpointer data,
GDestroyNotify destroy)
{
static guint seq_hook_id = 1;
GtkSignal *signal;
GHook *hook;
g_return_val_if_fail (signal_id > 0, 0);
g_return_val_if_fail (hook_func != NULL, 0);
signal = LOOKUP_SIGNAL_ID (signal_id);
g_return_val_if_fail (signal != NULL, 0);
if (signal->signal_flags & GTK_RUN_NO_HOOKS)
{
g_warning ("gtk_signal_add_emission_hook_full(): signal \"%s\" does not support emission hooks",
signal->name);
return 0;
}
if (!signal->hook_list)
{
signal->hook_list = g_new (GHookList, 1);
g_hook_list_init (signal->hook_list, sizeof (GHook));
}
hook = g_hook_alloc (signal->hook_list);
hook->data = data;
hook->func = hook_func;
hook->destroy = destroy;
signal->hook_list->seq_id = seq_hook_id;
g_hook_prepend (signal->hook_list, hook);
seq_hook_id = signal->hook_list->seq_id;
return hook->hook_id;
}
void
gtk_signal_remove_emission_hook (guint signal_id,
guint hook_id)
{
GtkSignal *signal;
g_return_if_fail (signal_id > 0);
g_return_if_fail (hook_id > 0);
signal = LOOKUP_SIGNAL_ID (signal_id);
g_return_if_fail (signal != NULL);
if (!signal->hook_list || !g_hook_destroy (signal->hook_list, hook_id))
g_warning ("gtk_signal_remove_emission_hook(): could not find hook (%u)", hook_id);
}
static gboolean
gtk_emission_hook_marshaller (GHook *hook,
gpointer data_p)
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkEmissionHookData *data = data_p;
GtkEmissionHook func;
func = hook->func;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (!GTK_OBJECT_DESTROYED (data->object))
return func (data->object, data->signal_id,
data->n_params, data->params,
hook->data);
else
return TRUE;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
static guint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_signal_connect_by_type (GtkObject *object,
guint signal_id,
1997-11-24 22:37:52 +00:00
GtkSignalFunc func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gpointer func_data,
1997-11-24 22:37:52 +00:00
GtkSignalDestroy destroy_func,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gint object_signal,
gint after,
gint no_marshal)
1997-11-24 22:37:52 +00:00
{
GtkObjectClass *class;
1997-11-24 22:37:52 +00:00
GtkHandler *handler;
gint found_it;
GtkSignal *signal;
1997-11-24 22:37:52 +00:00
g_return_val_if_fail (object != NULL, 0);
g_return_val_if_fail (object->klass != NULL, 0);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
signal = LOOKUP_SIGNAL_ID (signal_id);
1997-11-24 22:37:52 +00:00
/* Search through the signals for this object and make
* sure the one we are adding is valid. We need to perform
* the lookup on the objects parents as well. If it isn't
* valid then issue a warning and return.
fixed an assertment. Sat Jun 6 06:01:24 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_emitv): fixed an assertment. * gtk/makeenums.awk: a script to generate the GtkEnumValue arrays from, this should eventually be done by gentypeinfo.el somewhen. * gtk/gtkenumvalues.c: new generated file to hold GtkEnumValue arrays. * gtk/gtktypeutils.h: new function gtk_enum_values() to retrive all the enum values of an enum type. * gtk/gtk.defs: * gtk/gtkcurve.h: * gtk/gtkobject.h: * gtk/gtkprivate.h: * gtk/gtkwidget.h: * gtk/gtkenums.h: brought enum/flags definitions in sync, added a few more enum definitions for bindings and pattern matching. * some more macro and GtkType fixups in various places. * gdk/gdktypes.h (enum): added a new value GDK_AFTER_MASK, which is used as a key-release modifier for the binding system. Fri Jun 5 06:06:06 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenu.h (struct _GtkMenu): removed GList*children, since it was a stale list pointer that is already present in GtkMenuShell. * gtk/gtkmenushell.h (struct _GtkMenuShellClass): added a signal GtkMenuShell::selection_done which is emitted after the menu shell poped down again and all possible menu items have been activated. Thu Jun 4 02:20:42 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenushell.c (gtk_menu_shell_button_release): flush the x-queue before activation of the menuitem, so the menu is actually taken off the screen prior to any menu item activation. * gtk/gtkctree.c (gtk_ctree_get_row_data): allow function invokation for NULL nodes. * gtk/gtkwidget.h: * gtk/gtkwidget.c: new function gtk_widget_stop_accelerator to stop the emission of the "add-accelerator" signal on a widget. this is usefull to prevent accelerator installation on certain widgets. * gtk/gtknotebook.c (gtk_notebook_menu_item_create): keep the menu labels left justified, by setting their alignment. stop accelerator installation for the menu items, since we use dynamic menus. Wed Jun 3 06:41:22 1998 Tim Janik <timj@gtk.org> * gtk/gtkmenufactory.c: adaptions to use the new accel groups. people should *really* use GtkItemFactory. this is only for preserving source compatibility where possible, use of GtkMenuFactory is deprecated as of now. * gtk/gtkobject.h (gtk_object_class_add_user_signal): new function to create user signals of type GTK_RUN_NO_RECURSE. don't know why i missed this possibility when i added gtk_object_class_add_user_signal in late january. * gtk/gtkmain.c (gtk_init): ignore subsequent function calls. Sun May 31 07:31:09 1998 Tim Janik <timj@gtk.org> * gtk/gtkaccelgroup.h: * gtk/gtkaccelgroup.c: new implementation of the accelerator concept. * gtk/gtkaccellabel.h: * gtk/gtkaccellabel.c: new widget derived from GtkLabel whitch features display of the accelerators associated with a certain widget. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: new widget, item factory with automatic rc parsing and accelerator handling. * gtk/gtkmenu.c (gtk_menu_reposition): new function to care for positioning a menu. (gtk_menu_map): removed the allocation code. (gtk_menu_size_allocate): care for redrawing of children and resize our widget->window correctly. (gtk_menu_key_press): feature the new accelerator groups. * gtk/gtkmenuitem.c (gtk_menu_item_size_allocate): reposition the submenu if neccessary. * gtk/gtkmenuitem.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkradiomenuitem.c: use GtkAccelLabel in the *_new_with_label() function variants. * gdk/gdk.c: (gdk_keyval_from_name): (gdk_keyval_name): new functions for keyval<->key-name associations. (gdk_keyval_to_upper): (gdk_keyval_to_lower): (gdk_keyval_is_upper): (gdk_keyval_is_lower): new functions to check/translate keyvalues with regards to their cases. Wed May 27 00:48:10 1998 Tim Janik <timj@gtk.org> * gtk/gtkwidget.c (gtk_widget_class_path): new function to calculate a widget's class path. (gtk_widget_path): new function to calculate a widget's name path. * gtk/gtkrc.c: newly introduced GtkPatternSpec structures to speed up pattern matching, features reversed pattern matches.
1998-06-07 06:48:56 +00:00
* As of now (1998-05-27) this lookup shouldn't be neccessarry
* anymore since gtk_signal_lookup() has been reworked to only
* return correct signal ids per class-branch.
1997-11-24 22:37:52 +00:00
*/
found_it = FALSE;
class = object->klass;
while (class)
{
GtkType parent;
guint *object_signals;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
guint nsignals;
guint i;
object_signals = class->signals;
nsignals = class->nsignals;
for (i = 0; i < nsignals; i++)
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
if (object_signals[i] == signal_id)
{
found_it = TRUE;
break;
}
parent = gtk_type_parent (class->type);
if (parent)
class = gtk_type_class (parent);
else
class = NULL;
}
1997-11-24 22:37:52 +00:00
if (!found_it)
{
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
g_warning ("gtk_signal_connect_by_type(): could not find signal id (%u) in the `%s' class ancestry",
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
signal_id,
gtk_type_name (object->klass->type));
1997-11-24 22:37:52 +00:00
return 0;
}
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
handler = gtk_signal_handler_new ();
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
handler->id = gtk_handler_id++;
handler->signal_id = signal_id;
Tue Mar 16 17:43:33 1999 Tim Janik <timj@gtk.org> Wed Mar 17 01:46:28 1999 Tim Janik <timj@gtk.org> * merges from gtk-1-2: Tue Mar 16 17:43:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.c (gtk_item_factory_parse_rc_string): ensure the item factory class has been created. (gtk_item_factory_parse_rc): likewise. * gtk/gtkmenu.c: keep proper references for old_active_menu_item. (gtk_menu_reparent): unset the usize of the new parent, so the menu can sanely be size requested and we don't get nasty screen artefacts upon next reparentation. (gtk_menu_motion_notify): set send_event to TRUE if we synthesize an enter notify. only synthesize enter notifies if the pointer really is inside the event window. (gtk_menu_popdown): use gtk_menu_shell_deselect(). (gtk_menu_popup): move the background setting stuff into gtk_menu_tearoff_bg_copy() so it can be called from other places as well. * gtk/gtkmenushell.c (gtk_menu_shell_button_press): use gtk_menu_shell_select_item() to select the new item. (gtk_menu_shell_deselect): export this function, so gtkmenu.c can do the right thing for deselection as well. Sat Mar 15 20:10:33 1999 Tim Janik <timj@gtk.org> * gtk/gtkwidget.[hc]: (gtk_widget_accelerators_locked): return whether a widget's accelerators are locked. * gtk/gtkmenu.c (gtk_menu_key_press): don't remove or install new or existing accelerators if the widget's accelerators are locked. Sat Mar 14 19:44:05 1999 Tim Janik <timj@gtk.org> * gtk/gtkitemfactory.[hc]: allow managing of foreign menu items. * gtk/gtkmenu.c: truely forward key press and key release events to the menu widget from the toplevel or tearoff window. we can't simply connect to that, we need to stop further processing of the events as well. Sat Mar 13 13:14:17 1999 Tim Janik <timj@gtk.org> * gtk/gtkmenu.c: (gtk_menu_key_press): pass event->keyval, event->state to gtk_accelerator_valid, instead of event->keyval twice. refuse to install single letter accelerators for menus that use single letter shortcuts. * gtk/gtkitemfactory.c (gtk_item_factory_create_item): use gtk_menu_ensure_uline_accel_group(). * gtk/gtkmenu.[hc]: added gtk_menu_ensure_uline_accel_group() which will always return an uline accel group, made gtk_menu_get_uline_accel_group() return NULL if the group isn't yet created. Mon Mar 15 01:03:27 1999 Lars Hamann <lars@gtk.org> * gtk/gtkclist.h (struct _GtkCListColumn): added button_passive flag. * gtk/gtkclist.c (gtk_clist_column_title_passive): Leave button sensitive, trap button_press, button_release, motion_notify, enter_notify and leave_notify events instead. (gtk_clist_column_title_active): disconnect event handler. (gtk_clist_drag_data_get): fixed memory leak. Reported by Guillaume Laurent <glaurent@worldnet.fr> Wed Mar 10 23:49:55 1999 Lars Hamann <lars@gtk.org> * gtk/gtklayout.c (gtk_layout_adjustment_changed): fixed a few width/height mixups. * gtk/gtkctree.c (tree_delete): emit an tree_unselect_row signal if needed. Wed Mar 10 00:11:32 1999 Tim Janik <timj@gtk.org> * gtk/testgtk.c (create_item_factory): unref the item factory after window's destruction. * gtk/gtkmenushell.c (gtk_menu_shell_activate_item): keep a reference count on the menu shell around the menu item's activation, since the signal emission may cause menu shell destruction. * gtk/gtkitemfactory.c: the previous code leaked one accel group per menu. we use gtk_menu_get_uline_accel_group() now to fix that, and with that also create the underline accelerator group of the menus only if required (i.e. an underline accelerator has been specified). (gtk_item_factory_construct): (gtk_item_factory_create_item): removed code that would create an extra accel group for the menu (and leak references). (gtk_item_factory_create_item): adapted the underline accelerator installation code to properly feature gtk_menu_get_uline_accel_group(). * gtk/gtkmenu.[hc]: added gtk_menu_get_accel_group() to retrive menu->accel_group, this may return NULL if the accelerator group hasn't been set yet. added gtk_menu_get_uline_accel_group() to retrive the underline accelerator group of the menu, this will be created on demand and proper care is taken about its reference count. * gtk/gtkitemfactory.h: * gtk/gtkitemfactory.c: dumped the approach of keeping a widgets by action list on the factory since the factory<->widget destroy negotiation didn't work and would be hard to get going at all. instead we keep a list of GtkItemFactoryItem items on the factory (GtkItemFactoryItems are persistant throughout a program's life time). also, i removed the static const gchar *key_* variables, and made them inline strings (they weren't actually used anyways). (gtk_item_factory_add_item): update ifactory->items. (gtk_item_factory_destroy): destroy ifactory->items (and remove the item factory pointer from the remaining ifactory widgets). (gtk_item_factory_get_widget_by_action): walk the GtkItemFactoryItem list to find the widget. (gtk_item_factory_get_item): new function that works around gtk_item_factory_get_widget() limitations, this function will only return menu items, even for <Branch> entries. Tue Mar 9 01:01:28 1999 Tim Janik <timj@gtk.org> * gdk/gdkfont.c (gdk_font_load): first lookup the xfont ID in our font hash table, if we have a GdkFontPrivate entry for this font already, simply increment its reference count, provided by Olaf Dietsche <olaf.dietsche+list.gtk@netcologne.de>. * gtk/gtkstyle.c (gtk_style_copy): plug a GdkFont reference leak, fix provided by Olaf Dietsche <olaf.dietsche+list.gtk@netcologne.de>. Sun Mar 7 06:13:29 1999 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.c: (gtk_container_add_with_args): (gtk_container_addv): (gtk_container_add): before adding a child to a conatiner, make sure it is (default) constructed, this is neccessary because under certain circumstances the child will get relized and mapped immediatedly, in which case it has to be constructed already. Mon Mar 1 17:58:21 1999 Tim Janik <timj@gtk.org> * gtk/gtksignal.c (gtk_signal_connect_by_type): count object_signal values > 1 as TRUE also.
1999-03-17 01:39:42 +00:00
handler->object_signal = object_signal != FALSE;
1997-11-24 22:37:52 +00:00
handler->func = func;
handler->func_data = func_data;
handler->destroy_func = destroy_func;
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handler->after = after != FALSE;
1997-11-24 22:37:52 +00:00
handler->no_marshal = no_marshal;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
gtk_signal_handler_insert (object, handler);
return handler->id;
}
static GtkEmission*
configure.in acheader.h gdk/gdkwindow.c Check for Shape extension both on Sun May 3 13:38:22 1998 Owen Taylor <otaylor@gtk.org> * configure.in acheader.h gdk/gdkwindow.c Check for Shape extension both on the client and server side. (And, more importantly, check for the shape extension so we may include -lXext even when compiling with --disable-xshm) Don't set override_redirect on all shaped windows. It isn't necessary. * gdk/gdkwindow.c: Set ->colormap to NULL for root and foreign windows. Use this to check if we need to get the colormap from X. Fri May 1 22:32:47 1998 Owen Taylor <otaylor@gtk.org> * gtk/gtkbutton.c (gtk_button_paint): Draw the areas between the default and the button always in GTK_STATE_NORMAL. * gtk/gtkrange.c (gtk_range_style_set): Added a style_set callback. Fri May 1 16:40:57 1998 Owen Taylor <otaylor@gtk.org> * gdk/gdkpixmap.c (gdk_pixmap_colormap_create_from_xpmp[_d]): Fix a buffer overflow on pixmaps that claim to have more than 31 characters per pixel. (gdk_pixmap_read_string): Don't wrap around strings longer than half of address space ;-) * gtk/gtk[vh]ruler.c gtk/gtkinputdialog.c: Expand some buffers that were used for printing integers. * */* (almost): Style: All int foo () { ... } changed to int foo (void) { ... } ^^^^^^^ This is why some many files changed Even where there were proper prototypes elsewhere. * gdk/gxid.c (handle_claim_device): Some extra checks. It isn't safe against being fed bad X id's, but at least it should be safe against deleting all your files.
1998-05-03 22:41:32 +00:00
gtk_emission_new (void)
1997-11-24 22:37:52 +00:00
{
GtkEmission *emission;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (!gtk_free_emissions)
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkEmission *emission_block;
guint i;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission_block = g_new0 (GtkEmission, EMISSION_BLOCK_SIZE);
for (i = 1; i < EMISSION_BLOCK_SIZE; i++)
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
(emission_block + i)->next = gtk_free_emissions;
gtk_free_emissions = (emission_block + i);
}
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission = emission_block;
}
else
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission = gtk_free_emissions;
gtk_free_emissions = emission->next;
}
1997-11-24 22:37:52 +00:00
emission->object = NULL;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
emission->signal_id = 0;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission->in_hook = 0;
emission->next = NULL;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
return emission;
}
static void
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
gtk_emission_add (GtkEmission **emissions,
GtkObject *object,
guint signal_id)
1997-11-24 22:37:52 +00:00
{
GtkEmission *emission;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (emissions != NULL);
g_return_if_fail (object != NULL);
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
emission = gtk_emission_new ();
emission->object = object;
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
emission->signal_id = signal_id;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission->next = *emissions;
*emissions = emission;
1997-11-24 22:37:52 +00:00
}
static void
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
gtk_emission_remove (GtkEmission **emissions,
GtkObject *object,
guint signal_id)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
GtkEmission *emission, *last;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
1997-11-24 22:37:52 +00:00
g_return_if_fail (emissions != NULL);
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
last = NULL;
emission = *emissions;
while (emission)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (emission->object == object && emission->signal_id == signal_id)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (last)
last->next = emission->next;
else
*emissions = emission->next;
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
emission->next = gtk_free_emissions;
gtk_free_emissions = emission;
1997-11-24 22:37:52 +00:00
break;
}
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
last = emission;
emission = last->next;
1997-11-24 22:37:52 +00:00
}
}
static gint
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
gtk_emission_check (GtkEmission *emission,
GtkObject *object,
guint signal_id)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
while (emission)
1997-11-24 22:37:52 +00:00
{
reverted marius change to expose the type systems internal type info data Mon Dec 7 03:08:39 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h: * gtk/gtktypeutils.c: reverted marius change to expose the type systems internal type info data to the user. if such functionality is required we should provide wrapped accessors, ala gtk_signal_query(). * gtk/gtksignal.c (gtk_signal_connect_by_type): reverted marius change, since it destroys the possibility to implement automatic marshaller lookups some day, and it also disables third party code's ability to connect to any signal. also the GtkTypeInfo structures are dynamically allocated memory portions, so only the type system is really allowed to access that stuff. Mon Dec 7 01:32:18 1998 Tim Janik <timj@gtk.org> * gtk/gtkfilesel.c (gtk_file_selection_key_press): always intercept the Tab key on the entry. the focus shouldn't get lost even if completion is attempted from an empty entry, since an empty entry string does indeed have a valid completion meaning (complete all). (gtk_file_selection_init): cast the gchar array parameter in calls to gtk_clist_new_with_titles() to quit compiler warnings. (check_dir): the no_stat_dirs struct must not be const, since we do indeed modify its contents. * gtk/testgtk.c (event_watcher): adapted prototype to fit new emission hook semantics. * gtk/gtksignal.h: * gtk/gtksignal.c: changed emission allocation, so we don't use a doubly linked list but link ourselfs (singly linked). changed emission hooks, they get the emision parameters passed as well now and are emitted during the actuall signal emission (after the RUN_FIRST class method, but prior to RUN_FIRST handlers). the existing restrictions do still apply to signal emission hooks, i.e. an emission may not be stopped or restarted from an emission hook. due to possibly huge perfomance impacts, frequent use of emision hooks is also not recommended. (gtk_signal_next_and_invalidate): added an assertments which explicits what the code assumes anyways: a maximum amount of 65535 signals. * gtk/gtkcontainer.h: deprecated gtk_container_foreach_interp(), gtk_container_foreach_full() should be used instead. * gtk/gtkmain.h: deprecated gtk_timeout_add_interp and gtk_idle_add_interp, since we provide _full variants. * gtk/gtksignal.h: deprecated gtk_signal_connect_interp(), we provide gtk_signal_connect_full() for long enough now.
1998-12-07 02:31:19 +00:00
if (emission->object == object && emission->signal_id == signal_id)
return 1 + emission->in_hook;
emission = emission->next;
1997-11-24 22:37:52 +00:00
}
return FALSE;
}
static gint
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gtk_handlers_run (GtkHandler *handlers,
GtkSignal *signal,
GtkObject *object,
GtkArg *params,
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
gint after)
1997-11-24 22:37:52 +00:00
{
/* *signal is a local copy on the stack of gtk_signal_real_emit(),
* so we don't need to look it up every time we invoked a function.
*/
while (handlers && handlers->signal_id == signal->signal_id)
1997-11-24 22:37:52 +00:00
{
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
GtkHandler *handlers_next;
made the <widget>_signals[] arrays of type guint rather than gint. made Mon Mar 9 15:48:10 1998 Tim Janik <timj@gimp.org> * Signal signedness and naming corrections, plus GtkType fixes: * gtk/gtkadjustment.c: * gtk/gtkbutton.c: * gtk/gtkcheckmenuitem.c: * gtk/gtkclist.c: * gtk/gtkcolorsel.c: * gtk/gtkcontainer.c: * gtk/gtkcurve.c: * gtk/gtkdata.c: * gtk/gtkeditable.c: * gtk/gtkentry.c: * gtk/gtkhandlebox.c: * gtk/gtkinputdialog.c: * gtk/gtkitem.c: * gtk/gtklist.c: * gtk/gtkmenuitem.c: * gtk/gtkmenushell.c: * gtk/gtknotebook.c: * gtk/gtkstatusbar.c: * gtk/gtktoolbar.c: * gtk/gtktree.c: * gtk/gtktreeitem.c: * gtk/gtkwidget.c: * gtk/gtktogglebutton.c: * gtk/gtkwindow.c: made the <widget>_signals[] arrays of type guint rather than gint. * gtk/gtkwidget.c (gtk_widget_get_ancestor): made widget_type a GtkType. * gtk/gtkcombo.h: handler ids need to be of type guint (entry_change_id, list_change_id). * gtk/gtkaccelerator.c: changed signal_num to signal_id and typed it guint. * gtk/gtkmain.c: made gtk_ndebug_keys a guint. * gtk/gtkmenu.h: * gtk/gtkmenu.c: (gtk_menu_popup): made button a guint. (gtk_menu_set_active): made index a guint. * gtk/gtkmenuitem.h: * gtk/gtkmenuitem.c: made accelerator_signal a guint. * gtk/gtkoptionmenu.h: * gtk/gtkoptionmenu.c: (gtk_option_menu_set_history): made index a guint. * gtk/gtksignal.h: * gtk/gtksignal.c: * gtk/gtkobject.h: * gtk/gtkobject.c: changed a bunch of prototypes to take guints rather than gints. also made some conversions from guint to GtkType, left over from when the fundamental-types system was introduced. * gtk/gtkobject.h: * gtk/gtkobject.c: made object_data_id_index and obj_count guints. made *signals and nsignals guints in GtkObjectClass.
1998-03-09 15:16:28 +00:00
gtk_signal_handler_ref (handlers);
if (!handlers->blocked && handlers->after == after)
1997-11-24 22:37:52 +00:00
{
if (handlers->func)
{
if (handlers->no_marshal)
(* (GtkCallbackMarshal) handlers->func) (object,
handlers->func_data,
signal->nparams,
params);
1997-11-24 22:37:52 +00:00
else if (handlers->object_signal)
/* don't cast with GTK_OBJECT () */
(* signal->marshaller) ((GtkObject*) handlers->func_data,
handlers->func,
object,
params);
1997-11-24 22:37:52 +00:00
else
(* signal->marshaller) (object,
handlers->func,
handlers->func_data,
params);
1997-11-24 22:37:52 +00:00
}
else if (global_marshaller)
(* global_marshaller) (object,
handlers->func_data,
signal->nparams,
params,
signal->params,
signal->return_val);
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
if (stop_emissions && gtk_emission_check (stop_emissions,
object,
signal->signal_id))
1997-11-24 22:37:52 +00:00
{
gtk_emission_remove (&stop_emissions, object, signal->signal_id);
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
gtk_signal_handler_unref (handlers, object);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return EMISSION_DONE;
1997-11-24 22:37:52 +00:00
}
else if (restart_emissions &&
signal->signal_flags & GTK_RUN_NO_RECURSE &&
gtk_emission_check (restart_emissions, object, signal->signal_id))
1997-11-24 22:37:52 +00:00
{
gtk_emission_remove (&restart_emissions, object, signal->signal_id);
gtk_signal_handler_unref (handlers, object);
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return EMISSION_RESTART;
1997-11-24 22:37:52 +00:00
}
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
handlers_next = handlers->next;
gtk_signal_handler_unref (handlers, object);
handlers = handlers_next;
1997-11-24 22:37:52 +00:00
}
ok, there have been several severe bugs in the signal handler referencing Tue Feb 10 07:12:07 1998 Tim Janik <timj@gimp.org> * gtk/gtksignal.h: * gtk/gtksignal.c: ok, there have been several severe bugs in the signal handler referencing and ->next connection stuff. these bugs caused invokations of handlers that are disconnected and - worse - destroyed already. invokation of *destroyd* handlers mean: anything can be executed , because the handler structure can just as well be realocated. at the cost of an extra ->prev field per handler we should have a reasonable stable system now, because of the various places that can cause a handler to be disconnected (*any* handler invokation can cause *any* or *all* handlers to be disconnected, there is no way around a doubly linked list, actually handler disconnection has never worked correctly because of this. handlers are connected together via a *doubly* linked list now, and it is *not* valid to remove a handler out of this list untill all its references have been droped, i.e. handler->ref_count==0. to prevent emissions of disconnected but still referenced handlers, disconnected handlers are simply marked as blocked and get an id of 0 which is an invalid signal handler id. the handler->id has been changed to have 28 significant bits (using alignment gaps), since 65536 (old range: guint16) signal connections (as a total) can easily be reached by complex applications. this whole handler thingy is at least as tedious as writing doubly linked list implementations ;)
1998-02-10 06:53:08 +00:00
reimplemented the signal storage system to use a linear array rather than Wed May 6 02:16:34 1998 Tim Janik <timj@gtk.org> * gtk/gtksignal.c: reimplemented the signal storage system to use a linear array rather than a hash table. be carefull *not* to keep any GtkSignal pointers across invokations of gtk_signal_next_and_invalidate() and therefore gtk_signal_new[v](). general code cleanups, made all allocations through memchunks. (gtk_signal_lookup): we now do the lookup of signals through a key_id for the signal names to avoid multiple hashing of the signal name on lookups for the several hirarchy levels of an object. Tue May 5 15:33:27 1998 Tim Janik <timj@gtk.org> * gtk/gtkmain.c (gtk_init): added gtk_signal_init() to avoid repeatedly checking for signal initialization in gtksignal.c. * gtk/gtktypeutils.c (TYPE_NODES_BLOCK_SIZE): reseted this to 200 after a long debugging period ;) Mon May 4 00:30:11 1998 Tim Janik <timj@gtk.org> * gtk/gtktypeutils.h (GTK_TYPE_IS_A): new macro to speedup gtk_type_is_a(). * gtk/gtktypeutils.c: reimplemented the type storage system to use a linear array rather than a hash table. it actually speeded up testgtk for a *considerable* amount. be carefull *not* to keep any GtkTypeNode pointers across invokations of gtk_type_node_next_and_invalidate() and therefore gtk_type_unique()! * gtk/gtkobject.h (GTK_IS_OBJECT_CLASS): new macro to test for the inheritance of a class pointer. (GTK_OBJECT): modifications for speedups. (GTK_IS_OBJECT): likewise. (GTK_OBJECT_CLASS): likewise. (suggested by owen). * gtk/gtkcontainer.h (GTK_IS_CONTAINER_CLASS): new macro. * gtk/gtkwidget.h (GTK_IS_WIDGET_CLASS): new macro. * gtk/gtk.defs (GtkTooltips): define GtkTooltips as object not as boxed, since its derivation changed 4 month ago ;(. showed up with the unique type name checking code in gtk_type_unique(). * random guint->GtkType and macro fixups.
1998-05-06 01:43:56 +00:00
return EMISSION_CONTINUE;
1997-11-24 22:37:52 +00:00
}
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
static gboolean
gtk_signal_collect_params (GtkArg *params,
guint n_params,
GtkType *param_types,
GtkType return_type,
va_list var_args)
1997-11-24 22:37:52 +00:00
{
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
register GtkArg *last_param;
register gboolean failed = FALSE;
for (last_param = params + n_params; params < last_param; params++)
1997-11-24 22:37:52 +00:00
{
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
register gchar *error;
params->name = NULL;
params->type = *(param_types++);
GTK_ARG_COLLECT_VALUE (params,
var_args,
error);
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
if (error)
1997-11-24 22:37:52 +00:00
{
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
failed = TRUE;
g_warning ("gtk_signal_collect_params(): %s", error);
g_free (error);
1997-11-24 22:37:52 +00:00
}
}
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
params->type = return_type;
params->name = NULL;
return_type = GTK_FUNDAMENTAL_TYPE (return_type);
if (return_type != GTK_TYPE_NONE)
1997-11-24 22:37:52 +00:00
{
if ((return_type >= GTK_TYPE_FLAT_FIRST &&
return_type <= GTK_TYPE_FLAT_LAST) ||
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
(return_type == GTK_TYPE_OBJECT))
{
GTK_VALUE_POINTER (*params) = va_arg (var_args, gpointer);
if (GTK_VALUE_POINTER (*params) == NULL)
{
failed = TRUE;
g_warning ("gtk_signal_collect_params(): invalid NULL pointer for return argument type `%s'",
gtk_type_name (params->type));
}
}
else
{
failed = TRUE;
g_warning ("gtk_signal_collect_params(): unsupported return argument type `%s'",
gtk_type_name (params->type));
}
1997-11-24 22:37:52 +00:00
}
deleted most of the argument handling code, since that is now implemented Fri Jul 10 00:02:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkcontainer.h: * gtk/gtkcontainer.c: deleted most of the argument handling code, since that is now implemented in gtkarg.c. similar to gtk_object_args_collect, we now export a new function gtk_container_child_args_collect(). for consistency with the object arguments, a few functions got renamed: gtk_container_child_arg_get -> gtk_container_child_get, gtk_container_child_arg_set -> gtk_container_child_set, gtk_container_child_arg_getv -> gtk_container_child_getv, gtk_container_child_arg_setv -> gtk_container_child_setv, gtk_container_add_with_argv -> gtk_container_addv. note, gtk_container_add_with_args() remained, because its equivalent would be gtk_container_add(,, ...) which would break all existing code. (gtk_container_add_child_arg_type): similar to gtk_object_add_arg_type, we expect the `arg_name' argument to be a const static string now. (gtk_container_get_child_arg_type): function removed. Thu Jul 9 07:03:04 1998 Tim Janik <timj@gtk.org> * gtk/gtkargcollector.c: new file which holds gtk_arg_collect_value(). this is a static inline function that collects command line arguments from a va_list. this file can just be included in all places that need this functionality. * gtk/gtkarg.h: * gtk/gtkarg.c: new files which implement most of the argument handling stuff from gtkobject.c. also collected a few more gtk_arg_* utility functions from else places. * gtk/gtkobject.h: * gtk/gtkobject.c: moved most of the argument handling code into gtkarg.c. we now export gtk_object_args_collect() as a non-public method with a blind va_list pointer. (gtk_object_add_arg_type): the `arg_name' argument is required to be a const static string now. (gtk_object_get_arg_type): function got removed. * gtk/gtkwidget.c: (gtk_widget_set): (gtk_widget_new): adaptions for gtk_object_args_collect(). * gtk/gtktypeutils.c (gtk_type_init_builtin_types): changed the internal fundamental type name so as to have a valid prefix, e.g. "bool"-> "gboolean", "string"->"GtkString" and somesuch, left "void" as is, though that should probably be something like GtkNone since the type itself is called GTK_TYPE_NONE. even the internal type names need to avoid name clashes and must live in their own namespace, several code portions rely on that. we should relly have typedefs such as typedef gchar* GtkString; so the fundamental type names can be used for code dumpers just like with all the Gtk/Gdk types.
1998-07-09 23:00:34 +00:00
else
GTK_VALUE_POINTER (*params) = NULL;
return failed;
1997-11-24 22:37:52 +00:00
}