gtk2/gdk/win32/gdkwindow-win32.c

3615 lines
95 KiB
C
Raw Normal View History

/* GDK - The GIMP Drawing Kit
* Copyright (C) 1995-1997 Peter Mattis, Spencer Kimball and Josh MacDonald
* Copyright (C) 1998-2004 Tor Lillqvist
* Copyright (C) 2001-2004 Hans Breuer
* Copyright (C) 2007 Cody Russell
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser 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
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser 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.
*/
/*
* Modified by the GTK+ Team and others 1997-2000. 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/.
*/
#include <config.h>
#include <stdlib.h>
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
#include "gdk.h"
#include "gdkprivate-win32.h"
#include "gdkinput-win32.h"
#if 0
#include <gdk-pixbuf/gdk-pixbuf.h>
#include <stdio.h>
#endif
static GdkColormap* gdk_window_impl_win32_get_colormap (GdkDrawable *drawable);
static void gdk_window_impl_win32_set_colormap (GdkDrawable *drawable,
GdkColormap *cmap);
static void gdk_window_impl_win32_get_size (GdkDrawable *drawable,
gint *width,
gint *height);
static GdkRegion* gdk_window_impl_win32_get_visible_region (GdkDrawable *drawable);
static void gdk_window_impl_win32_init (GdkWindowImplWin32 *window);
static void gdk_window_impl_win32_class_init (GdkWindowImplWin32Class *klass);
static void gdk_window_impl_win32_finalize (GObject *object);
static gpointer parent_class = NULL;
static void update_style_bits (GdkWindow *window);
static gboolean _gdk_window_get_functions (GdkWindow *window,
GdkWMFunction *functions);
#define WINDOW_IS_TOPLEVEL(window) \
(GDK_WINDOW_TYPE (window) != GDK_WINDOW_CHILD && \
GDK_WINDOW_TYPE (window) != GDK_WINDOW_FOREIGN)
GType
_gdk_window_impl_win32_get_type (void)
{
static GType object_type = 0;
if (!object_type)
{
static const GTypeInfo object_info =
{
sizeof (GdkWindowImplWin32Class),
(GBaseInitFunc) NULL,
(GBaseFinalizeFunc) NULL,
(GClassInitFunc) gdk_window_impl_win32_class_init,
NULL, /* class_finalize */
NULL, /* class_data */
sizeof (GdkWindowImplWin32),
0, /* n_preallocs */
(GInstanceInitFunc) gdk_window_impl_win32_init,
};
object_type = g_type_register_static (GDK_TYPE_DRAWABLE_IMPL_WIN32,
"GdkWindowImplWin32",
&object_info, 0);
}
return object_type;
}
GType
_gdk_window_impl_get_type (void)
{
return _gdk_window_impl_win32_get_type ();
}
static void
gdk_window_impl_win32_init (GdkWindowImplWin32 *impl)
{
impl->width = 1;
impl->height = 1;
impl->toplevel_window_type = -1;
impl->hcursor = NULL;
impl->hicon_big = NULL;
impl->hicon_small = NULL;
impl->hint_flags = 0;
impl->type_hint = GDK_WINDOW_TYPE_HINT_NORMAL;
impl->extension_events_selected = FALSE;
impl->transient_owner = NULL;
}
Large changes to the Win32 backend, partially made necessary by the 2000-05-02 Tor Lillqvist <tml@iki.fi> Large changes to the Win32 backend, partially made necessary by the changes to the backend-independent internal structures. Attempts to implement similar backing store stuff as on X11. The current (CVS) version of the Win32 backend is *not* as stable as it was before the no-flicker branch was merged. A zipfile with that version is available from http://www.gimp.org/win32/. That should be use by "production" code until this CVS version is usable. (But note, the Win32 backend has never been claimed to be "production quality".) * README.win32: Add the above comment about versions. * gdk/gdkwindow.c: Don't use backing store for now on Win32. * gdk/gdk.def: Update. * gdk/gdkfont.h: Declare temporary Win32-only functions. Will presumably be replaced by some more better mechanism as 1.4 gets closer to release shape. * gdk/makefile.{cygwin,msc}: Update. * gdk/win32/*.c: Correct inclusions of the backend-specific and internal headers. Change code according to changes in these. Use gdk_drawable_*, not gdk_window_* where necessary. * gdk/win32/gdkdnd-win32.c: Use MISC selector for GDK_NOTE, not our old DND. * gdk/win32/gdkdrawable-win32.c (gdk_win32_draw_text): Don't try to interpret single characters as UTF-8. Thanks to Hans Breuer. Use correct function name in warning messages. * gdk/win32/gdkevents-win32.c: Use correct parameter lists for the GSourceFuncs gdk_event_prepare and gdk_event_check. (gdk_event_get_graphics_expose): Do implement, use PeekMessage. Thanks to Hans Breuer. (event_mask_string): Debugging function to print an GdkEventMask. (gdk_pointer_grab): Use it. * gdk/win32/gdkfont-win32.c: The Unicode subrange that the (old) book I used claimed was Hangul actually is CJK Unified Ideographs Extension A. Also, Hangul Syllables were missing. Improve logging. * gdk/win32/gdkgc-win32.c: Largish changes. * gdk/win32/gdkim-win32.c (gdk_set_locale): Use g_win32_getlocale() from GLib, and not setlocale() to get current locale name. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwin32.h: Move stuff from gdkprivate-win32.h to gdkwin32.h, similarily as in the X11 backend. * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Bugfix, assignment was used instead of equals in if test. Thanks to Hans Breuer. * gdk/win32/makefile.{cygwin,msc} * gtk/makefile.{cygwin,msc}: Updates. Better kludge to get the path to the Win32 headers that works also with the mingw compiler. * gtk/gtkstyle.c: Include <string.h>.
2000-05-01 22:06:49 +00:00
static void
gdk_window_impl_win32_class_init (GdkWindowImplWin32Class *klass)
{
GObjectClass *object_class = G_OBJECT_CLASS (klass);
GdkDrawableClass *drawable_class = GDK_DRAWABLE_CLASS (klass);
parent_class = g_type_class_peek_parent (klass);
object_class->finalize = gdk_window_impl_win32_finalize;
drawable_class->set_colormap = gdk_window_impl_win32_set_colormap;
drawable_class->get_colormap = gdk_window_impl_win32_get_colormap;
drawable_class->get_size = gdk_window_impl_win32_get_size;
/* Visible and clip regions are the same */
drawable_class->get_clip_region = gdk_window_impl_win32_get_visible_region;
drawable_class->get_visible_region = gdk_window_impl_win32_get_visible_region;
}
static void
gdk_window_impl_win32_finalize (GObject *object)
{
GdkWindowObject *wrapper;
GdkDrawableImplWin32 *draw_impl;
GdkWindowImplWin32 *window_impl;
g_return_if_fail (GDK_IS_WINDOW_IMPL_WIN32 (object));
draw_impl = GDK_DRAWABLE_IMPL_WIN32 (object);
window_impl = GDK_WINDOW_IMPL_WIN32 (object);
wrapper = (GdkWindowObject*) draw_impl->wrapper;
if (!GDK_WINDOW_DESTROYED (wrapper))
{
gdk_win32_handle_table_remove (draw_impl->handle);
}
if (window_impl->hcursor != NULL)
{
if (GetCursor () == window_impl->hcursor)
SetCursor (NULL);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDI_CALL (DestroyCursor, (window_impl->hcursor));
window_impl->hcursor = NULL;
}
if (window_impl->hicon_big != NULL)
{
GDI_CALL (DestroyIcon, (window_impl->hicon_big));
window_impl->hicon_big = NULL;
}
if (window_impl->hicon_small != NULL)
{
GDI_CALL (DestroyIcon, (window_impl->hicon_small));
window_impl->hicon_small = NULL;
}
G_OBJECT_CLASS (parent_class)->finalize (object);
}
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
void
_gdk_win32_adjust_client_rect (GdkWindow *window,
RECT *rect)
{
LONG style, exstyle;
style = GetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE);
exstyle = GetWindowLong (GDK_WINDOW_HWND (window), GWL_EXSTYLE);
API_CALL (AdjustWindowRectEx, (rect, style, FALSE, exstyle));
}
static GdkColormap*
gdk_window_impl_win32_get_colormap (GdkDrawable *drawable)
{
GdkDrawableImplWin32 *drawable_impl;
g_return_val_if_fail (GDK_IS_WINDOW_IMPL_WIN32 (drawable), NULL);
drawable_impl = GDK_DRAWABLE_IMPL_WIN32 (drawable);
if (!((GdkWindowObject *) drawable_impl->wrapper)->input_only &&
drawable_impl->colormap == NULL)
{
drawable_impl->colormap = gdk_screen_get_system_colormap (_gdk_screen);
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
g_object_ref (drawable_impl->colormap);
}
return drawable_impl->colormap;
}
static void
gdk_window_impl_win32_set_colormap (GdkDrawable *drawable,
GdkColormap *cmap)
{
GdkWindowImplWin32 *impl;
GdkDrawableImplWin32 *draw_impl;
g_return_if_fail (GDK_IS_WINDOW_IMPL_WIN32 (drawable));
impl = GDK_WINDOW_IMPL_WIN32 (drawable);
draw_impl = GDK_DRAWABLE_IMPL_WIN32 (drawable);
/* chain up */
GDK_DRAWABLE_CLASS (parent_class)->set_colormap (drawable, cmap);
if (cmap)
{
/* XXX */
g_print ("gdk_window_impl_win32_set_colormap: XXX\n");
}
}
static void
gdk_window_impl_win32_get_size (GdkDrawable *drawable,
gint *width,
gint *height)
{
g_return_if_fail (GDK_IS_WINDOW_IMPL_WIN32 (drawable));
if (width)
*width = GDK_WINDOW_IMPL_WIN32 (drawable)->width;
if (height)
*height = GDK_WINDOW_IMPL_WIN32 (drawable)->height;
}
static GdkRegion*
gdk_window_impl_win32_get_visible_region (GdkDrawable *drawable)
{
GdkWindowImplWin32 *impl = GDK_WINDOW_IMPL_WIN32 (drawable);
GdkRectangle result_rect;
result_rect.x = 0;
result_rect.y = 0;
result_rect.width = impl->width;
result_rect.height = impl->height;
gdk_rectangle_intersect (&result_rect, &impl->position_info.clip_rect, &result_rect);
return gdk_region_rectangle (&result_rect);
}
void
_gdk_root_window_size_init (void)
{
GdkWindowImplWin32 *impl;
GdkRectangle rect;
int i;
impl = GDK_WINDOW_IMPL_WIN32 (((GdkWindowObject *) _gdk_root)->impl);
rect = _gdk_monitors[0];
for (i = 1; i < _gdk_num_monitors; i++)
gdk_rectangle_union (&rect, _gdk_monitors+i, &rect);
impl->width = rect.width;
impl->height = rect.height;
}
void
_gdk_windowing_window_init (void)
{
GdkWindowObject *private;
GdkDrawableImplWin32 *draw_impl;
g_assert (_gdk_root == NULL);
_gdk_root = g_object_new (GDK_TYPE_WINDOW, NULL);
private = (GdkWindowObject *)_gdk_root;
draw_impl = GDK_DRAWABLE_IMPL_WIN32 (private->impl);
draw_impl->handle = GetDesktopWindow ();
draw_impl->wrapper = GDK_DRAWABLE (private);
draw_impl->colormap = gdk_screen_get_default_colormap (_gdk_screen);
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
g_object_ref (draw_impl->colormap);
private->window_type = GDK_WINDOW_ROOT;
private->depth = gdk_visual_get_system ()->depth;
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
_gdk_root_window_size_init ();
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
_gdk_window_init_position (GDK_WINDOW (private));
gdk_win32_handle_table_insert ((HANDLE *) &draw_impl->handle, _gdk_root);
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
GDK_NOTE (MISC, g_print ("_gdk_root=%p\n", GDK_WINDOW_HWND (_gdk_root)));
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
}
static const gchar *
get_default_title (void)
{
const char *title;
title = g_get_application_name ();
if (!title)
title = g_get_prgname ();
return title;
}
/* RegisterGdkClass
* is a wrapper function for RegisterWindowClassEx.
* It creates at least one unique class for every
* GdkWindowType. If support for single window-specific icons
* is ever needed (e.g Dialog specific), every such window should
* get its own class
*/
static ATOM
RegisterGdkClass (GdkWindowType wtype, GdkWindowTypeHint wtype_hint)
{
static ATOM klassTOPLEVEL = 0;
static ATOM klassDIALOG = 0;
static ATOM klassCHILD = 0;
static ATOM klassTEMP = 0;
static ATOM klassTEMPSHADOW = 0;
static HICON hAppIcon = NULL;
static HICON hAppIconSm = NULL;
static WNDCLASSEXW wcl;
ATOM klass = 0;
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
wcl.cbSize = sizeof (WNDCLASSEX);
wcl.style = 0; /* DON'T set CS_<H,V>REDRAW. It causes total redraw
* on WM_SIZE and WM_MOVE. Flicker, Performance!
*/
wcl.lpfnWndProc = _gdk_win32_window_procedure;
wcl.cbClsExtra = 0;
wcl.cbWndExtra = 0;
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
wcl.hInstance = _gdk_app_hmodule;
wcl.hIcon = 0;
wcl.hIconSm = 0;
/* initialize once! */
if (0 == hAppIcon && 0 == hAppIconSm)
{
gchar sLoc [MAX_PATH+1];
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
if (0 != GetModuleFileName (_gdk_app_hmodule, sLoc, MAX_PATH))
{
ExtractIconEx (sLoc, 0, &hAppIcon, &hAppIconSm, 1);
if (0 == hAppIcon && 0 == hAppIconSm)
{
if (0 != GetModuleFileName (_gdk_dll_hinstance, sLoc, MAX_PATH))
ExtractIconEx (sLoc, 0, &hAppIcon, &hAppIconSm, 1);
}
}
if (0 == hAppIcon && 0 == hAppIconSm)
{
hAppIcon = LoadImage (NULL, IDI_APPLICATION, IMAGE_ICON,
GetSystemMetrics (SM_CXICON),
GetSystemMetrics (SM_CYICON), 0);
hAppIconSm = LoadImage (NULL, IDI_APPLICATION, IMAGE_ICON,
GetSystemMetrics (SM_CXSMICON),
GetSystemMetrics (SM_CYSMICON), 0);
}
}
if (0 == hAppIcon)
hAppIcon = hAppIconSm;
else if (0 == hAppIconSm)
hAppIconSm = hAppIcon;
wcl.lpszMenuName = NULL;
/* initialize once per class */
/*
* HB: Setting the background brush leads to flicker, because we
* don't get asked how to clear the background. This is not what
* we want, at least not for input_only windows ...
*/
#define ONCE_PER_CLASS() \
wcl.hIcon = CopyIcon (hAppIcon); \
wcl.hIconSm = CopyIcon (hAppIconSm); \
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
wcl.hbrBackground = NULL; \
wcl.hCursor = LoadCursor (NULL, IDC_ARROW);
switch (wtype)
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
{
case GDK_WINDOW_TOPLEVEL:
if (0 == klassTOPLEVEL)
{
wcl.lpszClassName = L"gdkWindowToplevel";
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
ONCE_PER_CLASS ();
klassTOPLEVEL = RegisterClassExW (&wcl);
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
}
klass = klassTOPLEVEL;
break;
case GDK_WINDOW_CHILD:
if (0 == klassCHILD)
{
wcl.lpszClassName = L"gdkWindowChild";
wcl.style |= CS_PARENTDC; /* MSDN: ... enhances system performance. */
ONCE_PER_CLASS ();
klassCHILD = RegisterClassExW (&wcl);
}
klass = klassCHILD;
break;
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
case GDK_WINDOW_DIALOG:
if (0 == klassDIALOG)
{
wcl.lpszClassName = L"gdkWindowDialog";
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
wcl.style |= CS_SAVEBITS;
ONCE_PER_CLASS ();
klassDIALOG = RegisterClassExW (&wcl);
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
}
klass = klassDIALOG;
break;
case GDK_WINDOW_TEMP:
if ((wtype_hint == GDK_WINDOW_TYPE_HINT_MENU) ||
(wtype_hint == GDK_WINDOW_TYPE_HINT_DROPDOWN_MENU) ||
(wtype_hint == GDK_WINDOW_TYPE_HINT_POPUP_MENU) ||
(wtype_hint == GDK_WINDOW_TYPE_HINT_TOOLTIP))
{
if (klassTEMPSHADOW == 0)
{
wcl.lpszClassName = L"gdkWindowTempShadow";
wcl.style |= CS_SAVEBITS;
if (LOBYTE (g_win32_get_windows_version()) > 0x05 ||
LOWORD (g_win32_get_windows_version()) == 0x0105)
{
/* Windows XP (5.1) or above */
wcl.style |= 0x00020000; /* CS_DROPSHADOW */
}
ONCE_PER_CLASS ();
klassTEMPSHADOW = RegisterClassExW (&wcl);
}
klass = klassTEMPSHADOW;
}
else
{
if (klassTEMP == 0)
{
wcl.lpszClassName = L"gdkWindowTemp";
wcl.style |= CS_SAVEBITS;
ONCE_PER_CLASS ();
klassTEMP = RegisterClassExW (&wcl);
}
klass = klassTEMP;
}
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
break;
default:
g_assert_not_reached ();
break;
}
if (klass == 0)
{
WIN32_API_FAILED ("RegisterClassExW");
g_error ("That is a fatal error");
}
return klass;
}
static GdkWindow*
gdk_window_new_internal (GdkWindow *parent,
GdkWindowAttr *attributes,
gint attributes_mask,
gboolean from_set_skip_taskbar_hint)
{
HWND hwndNew;
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
HANDLE hparent;
ATOM klass = 0;
DWORD dwStyle = 0, dwExStyle;
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
RECT rect;
GdkWindow *window;
GdkWindow *orig_parent;
GdkWindowObject *private;
GdkWindowImplWin32 *impl;
GdkDrawableImplWin32 *draw_impl;
GdkVisual *visual;
const gchar *title;
wchar_t *wtitle;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
gint window_width, window_height;
gint offset_x = 0, offset_y = 0;
g_return_val_if_fail (attributes != NULL, NULL);
if (!parent)
parent = _gdk_root;
g_return_val_if_fail (GDK_IS_WINDOW (parent), NULL);
orig_parent = parent;
GDK_NOTE (MISC,
g_print ("gdk_window_new: %s\n",
(attributes->window_type == GDK_WINDOW_TOPLEVEL ? "TOPLEVEL" :
(attributes->window_type == GDK_WINDOW_CHILD ? "CHILD" :
(attributes->window_type == GDK_WINDOW_DIALOG ? "DIALOG" :
(attributes->window_type == GDK_WINDOW_TEMP ? "TEMP" :
"???"))))));
if (GDK_WINDOW_DESTROYED (parent))
return NULL;
hparent = GDK_WINDOW_HWND (parent);
window = g_object_new (GDK_TYPE_WINDOW, NULL);
private = (GdkWindowObject *)window;
impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
draw_impl = GDK_DRAWABLE_IMPL_WIN32 (private->impl);
draw_impl->wrapper = GDK_DRAWABLE (window);
/* Windows with a foreign parent are treated as if they are children
* of the root window, except for actual creation.
*/
if (GDK_WINDOW_TYPE (parent) == GDK_WINDOW_FOREIGN)
parent = _gdk_root;
private->parent = (GdkWindowObject *)parent;
private->accept_focus = TRUE;
private->focus_on_map = TRUE;
if (attributes_mask & GDK_WA_X)
private->x = attributes->x;
else
private->x = 0;
if (attributes_mask & GDK_WA_Y)
private->y = attributes->y;
else if (attributes_mask & GDK_WA_X)
private->y = 100; /* ??? We must put it somewhere... */
else
private->y = 0;
if (attributes_mask & GDK_WA_VISUAL)
visual = attributes->visual;
else
visual = gdk_visual_get_system ();
impl->width = (attributes->width > 1) ? (attributes->width) : (1);
impl->height = (attributes->height > 1) ? (attributes->height) : (1);
impl->extension_events_selected = FALSE;
if (attributes->wclass == GDK_INPUT_ONLY)
{
/* Backwards compatiblity - we've always ignored
* attributes->window_type for input-only windows
* before
*/
if (parent == _gdk_root)
private->window_type = GDK_WINDOW_TEMP;
else
private->window_type = GDK_WINDOW_CHILD;
}
else
private->window_type = attributes->window_type;
if (attributes->wclass == GDK_INPUT_OUTPUT)
{
dwExStyle = 0;
private->input_only = FALSE;
private->depth = visual->depth;
if (attributes_mask & GDK_WA_COLORMAP)
{
draw_impl->colormap = attributes->colormap;
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
g_object_ref (attributes->colormap);
}
else
{
draw_impl->colormap = gdk_screen_get_system_colormap (_gdk_screen);
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
g_object_ref (draw_impl->colormap);
}
}
else
{
/* I very much doubt using WS_EX_TRANSPARENT actually
* corresponds to how X11 InputOnly windows work, but it appears
* to work well enough for the actual use cases in gtk.
*/
dwExStyle = WS_EX_TRANSPARENT;
private->depth = 0;
private->input_only = TRUE;
draw_impl->colormap = gdk_screen_get_system_colormap (_gdk_screen);
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
g_object_ref (draw_impl->colormap);
GDK_NOTE (MISC, g_print ("... GDK_INPUT_ONLY, system colormap"));
}
switch (private->window_type)
{
case GDK_WINDOW_TOPLEVEL:
case GDK_WINDOW_DIALOG:
if (parent != _gdk_root)
{
g_warning (G_STRLOC ": Toplevel windows must be created as children\n"
"of a window of type GDK_WINDOW_ROOT or GDK_WINDOW_FOREIGN");
hparent = GetDesktopWindow ();
}
/* Children of foreign windows aren't toplevel windows */
if (GDK_WINDOW_TYPE (orig_parent) == GDK_WINDOW_FOREIGN)
{
dwStyle = WS_CHILDWINDOW | WS_CLIPCHILDREN;
}
else
{
if (private->window_type == GDK_WINDOW_TOPLEVEL)
dwStyle = WS_OVERLAPPEDWINDOW | WS_CLIPCHILDREN;
else
dwStyle = WS_OVERLAPPED | WS_MINIMIZEBOX | WS_SYSMENU | WS_CAPTION | WS_THICKFRAME | WS_CLIPCHILDREN;
offset_x = _gdk_offset_x;
offset_y = _gdk_offset_y;
}
break;
case GDK_WINDOW_CHILD:
dwStyle = WS_CHILDWINDOW | WS_CLIPCHILDREN | WS_CLIPSIBLINGS;
break;
case GDK_WINDOW_TEMP:
/* A temp window is not necessarily a top level window */
dwStyle = (_gdk_root == parent ? WS_POPUP : WS_CHILDWINDOW);
dwStyle |= WS_CLIPCHILDREN | WS_CLIPSIBLINGS;
dwExStyle |= WS_EX_TOOLWINDOW;
offset_x = _gdk_offset_x;
offset_y = _gdk_offset_y;
break;
case GDK_WINDOW_ROOT:
g_error ("cannot make windows of type GDK_WINDOW_ROOT");
break;
default:
g_assert_not_reached ();
}
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
_gdk_window_init_position (GDK_WINDOW (private));
if (private->window_type != GDK_WINDOW_CHILD)
{
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
rect.left = rect.top = 0;
rect.right = impl->position_info.width;
rect.bottom = impl->position_info.height;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
AdjustWindowRectEx (&rect, dwStyle, FALSE, dwExStyle);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
window_width = rect.right - rect.left;
window_height = rect.bottom - rect.top;
}
else
{
window_width = impl->position_info.width;
window_height = impl->position_info.height;
}
if (attributes_mask & GDK_WA_TITLE)
title = attributes->title;
else
title = get_default_title ();
if (!title || !*title)
title = "";
private->event_mask = GDK_STRUCTURE_MASK | attributes->event_mask;
if (attributes_mask & GDK_WA_TYPE_HINT)
impl->type_hint = attributes->type_hint;
else
impl->type_hint = GDK_WINDOW_TYPE_HINT_NORMAL;
if (private->parent)
private->parent->children = g_list_prepend (private->parent->children, window);
klass = RegisterGdkClass (private->window_type, impl->type_hint);
wtitle = g_utf8_to_utf16 (title, -1, NULL, NULL, NULL);
hwndNew = CreateWindowExW (dwExStyle,
MAKEINTRESOURCEW (klass),
wtitle,
dwStyle,
((attributes_mask & GDK_WA_X) ?
impl->position_info.x - offset_x : CW_USEDEFAULT),
impl->position_info.y - offset_y,
window_width, window_height,
hparent,
NULL,
_gdk_app_hmodule,
window);
if (GDK_WINDOW_HWND (window) != hwndNew)
{
g_warning ("gdk_window_new: gdk_event_translate::WM_CREATE (%p, %p) HWND mismatch.",
GDK_WINDOW_HWND (window),
hwndNew);
/* HB: IHMO due to a race condition the handle was increased by
* one, which causes much trouble. Because I can't find the
* real bug, try to workaround it ...
* To reproduce: compile with MSVC 5, DEBUG=1
*/
# if 0
gdk_win32_handle_table_remove (GDK_WINDOW_HWND (window));
GDK_WINDOW_HWND (window) = hwndNew;
gdk_win32_handle_table_insert (&GDK_WINDOW_HWND (window), window);
# else
/* the old behaviour, but with warning */
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
draw_impl->handle = hwndNew;
# endif
}
g_object_ref (window);
gdk_win32_handle_table_insert (&GDK_WINDOW_HWND (window), window);
GDK_NOTE (MISC, g_print ("... \"%s\" %dx%d@%+d%+d %p = %p\n",
title,
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
window_width, window_height,
((attributes_mask & GDK_WA_X) ?
impl->position_info.x - offset_x: CW_USEDEFAULT),
impl->position_info.y - offset_y,
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
hparent,
GDK_WINDOW_HWND (window)));
g_free (wtitle);
if (draw_impl->handle == NULL)
{
WIN32_API_FAILED ("CreateWindowExW");
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
g_object_unref (window);
return NULL;
}
if (!from_set_skip_taskbar_hint && private->window_type == GDK_WINDOW_TEMP)
gdk_window_set_skip_taskbar_hint (window, TRUE);
gdk_window_set_cursor (window, ((attributes_mask & GDK_WA_CURSOR) ?
(attributes->cursor) :
NULL));
return window;
}
GdkWindow*
gdk_window_new (GdkWindow *parent,
GdkWindowAttr *attributes,
gint attributes_mask)
{
return gdk_window_new_internal (parent, attributes, attributes_mask, FALSE);
}
GdkWindow *
Changes multihead reorganizing code for win32 support, mostly from a patch Wed Jun 5 18:34:47 2002 Owen Taylor <otaylor@redhat.com> Changes multihead reorganizing code for win32 support, mostly from a patch by Hans Breuer. * gdk/gdkcolor.c gdk/x11/gdkcolor-x11.c gdk/gdkcursor.c gdk/x11/gdkcursor-x11.c gdk/gdkevents.c gdk/x11/gdkevents-x11.c gdk/gdkfont.c gdk/x11/gdkfont-x11.c gdk/gdkkeys.c gdk/x11/gdkkeys-x11.c gdk/gdkimage.c gdk/x11/gdkimage-x11.c gdk/gdkscreen.c gdk/x11/gdkmain-x11.c gdk/gdkdisplay.c gdk/gdkevents-x11.c gdk/gdkpango.c gdk/x11/gdkpango-x11.c gdk/gdkselection.c gdk/x11/gdkselection-x11.c gdk/gdkwindow.c gdk/x11/gdkwindow-x11.c gdk/gdkvisual.c gdk/x11/gdkvisual-x11.c: Move port-independent singlehead wrapper functions into port-independent part of GDK. (#80009) * gdk/win32/gdkcolor-win32.c gdk/win32/gdkcursor-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkfont-win32.c gdk/win32/gdkimage-win32.c gdk/win32/gdkkeys-win32.c gdk/win32/gdkmain-win32.c gdk/win32/gdkproperty-win32.c gdk/win32/gdkselection-win32.c gdk/win32/gkwindow-win32.c: Turn singlehead functions into "multihead" functions that ignore their GdkDisplay or GdkScreen arguments. * gdk/win32/gdkdrawable-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkinput-win32.c gdk/win32/gdkprivate-win32.h: Misc multihead-compatibility changes. * gtk/gtk.def gdk/gdk.def: Update for multihead functions. * gdk/gdkcolormap.h gdk/gdkvisual.h gdk/x11/gdkcolormap-x11.c gdk/x11/gdkvisual-x11.c: Remove the screen fields from the public parts of the colormap/visual structures, add accessors instead. * gdk/gdkpixbuf-render.c gdk/gdkpixmap.c gdk/gdkrgb.c gdk/x11/gdkcolormap-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkprivate-x11.h gtk/gtkgc.c gtk/gtkstyle.c gtk/gtkwidget.c: Use accessors to get the screen for colormaps, visuals; move the fields into the private structures for the x11 backend. * gdk/gdkdisplay.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/gdkscreen.[ch] gdk/x11/gdkscreen-x11.c: Remove virtualization of screen and display functions. (#79990, patch from Erwann Chenede) * gdk/win32/gdkdisplay-x11.c gdk/win32/gdkscreen-win32.c gdk/win32/{Makefile.am, makefile.msc, makefile.mingw}: New files containing stub implementations of Display, Screen functions. * gdk/x11/gdkscreen-x11.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/x11/gdkx.h: Clean up function exports and what headers they are in. (#79954) * gdk/x11/gdkx.h: Fix macro that was referring to a non-existant screen->screen_num. (In the patch for #79972, Erwann Chenede) * gdk/gdkscreen.c gdk/gdkwindow.c gdk/x11/gdkinternals.h gdk/x11/gdkscreen-x11.c: Fix gdk_screen_get_window_at_pointer() to use window hooks. (#79972, patch partly from Erwann Chenede) * gdk/x11/gdkdisplay-x11.c gdk/x11/gdkevents-x11.c: Fix some warnings.
2002-06-06 00:26:42 +00:00
gdk_window_foreign_new_for_display (GdkDisplay *display,
GdkNativeWindow anid)
{
GdkWindow *window;
GdkWindowObject *private;
GdkWindowImplWin32 *impl;
GdkDrawableImplWin32 *draw_impl;
HANDLE parent;
RECT rect;
POINT point;
g_return_val_if_fail (display == _gdk_display, NULL);
Changes multihead reorganizing code for win32 support, mostly from a patch Wed Jun 5 18:34:47 2002 Owen Taylor <otaylor@redhat.com> Changes multihead reorganizing code for win32 support, mostly from a patch by Hans Breuer. * gdk/gdkcolor.c gdk/x11/gdkcolor-x11.c gdk/gdkcursor.c gdk/x11/gdkcursor-x11.c gdk/gdkevents.c gdk/x11/gdkevents-x11.c gdk/gdkfont.c gdk/x11/gdkfont-x11.c gdk/gdkkeys.c gdk/x11/gdkkeys-x11.c gdk/gdkimage.c gdk/x11/gdkimage-x11.c gdk/gdkscreen.c gdk/x11/gdkmain-x11.c gdk/gdkdisplay.c gdk/gdkevents-x11.c gdk/gdkpango.c gdk/x11/gdkpango-x11.c gdk/gdkselection.c gdk/x11/gdkselection-x11.c gdk/gdkwindow.c gdk/x11/gdkwindow-x11.c gdk/gdkvisual.c gdk/x11/gdkvisual-x11.c: Move port-independent singlehead wrapper functions into port-independent part of GDK. (#80009) * gdk/win32/gdkcolor-win32.c gdk/win32/gdkcursor-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkfont-win32.c gdk/win32/gdkimage-win32.c gdk/win32/gdkkeys-win32.c gdk/win32/gdkmain-win32.c gdk/win32/gdkproperty-win32.c gdk/win32/gdkselection-win32.c gdk/win32/gkwindow-win32.c: Turn singlehead functions into "multihead" functions that ignore their GdkDisplay or GdkScreen arguments. * gdk/win32/gdkdrawable-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkinput-win32.c gdk/win32/gdkprivate-win32.h: Misc multihead-compatibility changes. * gtk/gtk.def gdk/gdk.def: Update for multihead functions. * gdk/gdkcolormap.h gdk/gdkvisual.h gdk/x11/gdkcolormap-x11.c gdk/x11/gdkvisual-x11.c: Remove the screen fields from the public parts of the colormap/visual structures, add accessors instead. * gdk/gdkpixbuf-render.c gdk/gdkpixmap.c gdk/gdkrgb.c gdk/x11/gdkcolormap-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkprivate-x11.h gtk/gtkgc.c gtk/gtkstyle.c gtk/gtkwidget.c: Use accessors to get the screen for colormaps, visuals; move the fields into the private structures for the x11 backend. * gdk/gdkdisplay.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/gdkscreen.[ch] gdk/x11/gdkscreen-x11.c: Remove virtualization of screen and display functions. (#79990, patch from Erwann Chenede) * gdk/win32/gdkdisplay-x11.c gdk/win32/gdkscreen-win32.c gdk/win32/{Makefile.am, makefile.msc, makefile.mingw}: New files containing stub implementations of Display, Screen functions. * gdk/x11/gdkscreen-x11.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/x11/gdkx.h: Clean up function exports and what headers they are in. (#79954) * gdk/x11/gdkx.h: Fix macro that was referring to a non-existant screen->screen_num. (In the patch for #79972, Erwann Chenede) * gdk/gdkscreen.c gdk/gdkwindow.c gdk/x11/gdkinternals.h gdk/x11/gdkscreen-x11.c: Fix gdk_screen_get_window_at_pointer() to use window hooks. (#79972, patch partly from Erwann Chenede) * gdk/x11/gdkdisplay-x11.c gdk/x11/gdkevents-x11.c: Fix some warnings.
2002-06-06 00:26:42 +00:00
window = g_object_new (GDK_TYPE_WINDOW, NULL);
private = (GdkWindowObject *)window;
impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
draw_impl = GDK_DRAWABLE_IMPL_WIN32 (private->impl);
draw_impl->wrapper = GDK_DRAWABLE (window);
parent = GetParent ((HWND)anid);
private->parent = gdk_win32_handle_table_lookup ((GdkNativeWindow) parent);
if (!private->parent || GDK_WINDOW_TYPE (private->parent) == GDK_WINDOW_FOREIGN)
private->parent = (GdkWindowObject *)_gdk_root;
private->parent->children = g_list_prepend (private->parent->children, window);
draw_impl->handle = (HWND) anid;
GetClientRect ((HWND) anid, &rect);
point.x = rect.left;
point.y = rect.right;
ClientToScreen ((HWND) anid, &point);
if (parent != GetDesktopWindow ())
ScreenToClient (parent, &point);
private->x = point.x;
private->y = point.y;
impl->width = rect.right - rect.left;
impl->height = rect.bottom - rect.top;
private->window_type = GDK_WINDOW_FOREIGN;
private->destroyed = FALSE;
private->event_mask = GDK_ALL_EVENTS_MASK; /* XXX */
if (IsWindowVisible ((HWND) anid))
private->state &= (~GDK_WINDOW_STATE_WITHDRAWN);
else
private->state |= GDK_WINDOW_STATE_WITHDRAWN;
if (GetWindowLong ((HWND)anid, GWL_EXSTYLE) & WS_EX_TOPMOST)
private->state |= GDK_WINDOW_STATE_ABOVE;
else
private->state &= (~GDK_WINDOW_STATE_ABOVE);
private->state &= (~GDK_WINDOW_STATE_BELOW);
private->depth = gdk_visual_get_system ()->depth;
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
_gdk_window_init_position (GDK_WINDOW (private));
g_object_ref (window);
gdk_win32_handle_table_insert (&GDK_WINDOW_HWND (window), window);
GDK_NOTE (MISC, g_print ("gdk_window_foreign_new_for_display: %p: %s@%+d%+d\n",
(HWND) anid,
_gdk_win32_drawable_description (window),
private->x, private->y));
return window;
}
GdkWindow*
gdk_window_lookup (GdkNativeWindow hwnd)
{
return (GdkWindow*) gdk_win32_handle_table_lookup (hwnd);
}
void
_gdk_windowing_window_destroy (GdkWindow *window,
gboolean recursing,
gboolean foreign_destroy)
{
GdkWindowObject *private = (GdkWindowObject *)window;
GdkWindowImplWin32 *window_impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
g_return_if_fail (GDK_IS_WINDOW (window));
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("_gdk_windowing_window_destroy: %p\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window)));
if (private->extension_events != 0)
_gdk_input_window_destroy (window);
/* Remove ourself from our transient owner */
if (window_impl->transient_owner != NULL)
{
gdk_window_set_transient_for (window, NULL);
}
if (!recursing && !foreign_destroy)
{
_gdk_win32_drawable_finish (private->impl);
private->destroyed = TRUE;
DestroyWindow (GDK_WINDOW_HWND (window));
}
gdk_win32_handle_table_remove (GDK_WINDOW_HWND (window));
}
void
_gdk_windowing_window_destroy_foreign (GdkWindow *window)
{
/* It's somebody else's window, but in our hierarchy, so reparent it
* to the desktop, and then try to destroy it.
*/
gdk_window_hide (window);
gdk_window_reparent (window, NULL, 0, 0);
PostMessage (GDK_WINDOW_HWND (window), WM_CLOSE, 0, 0);
}
/* This function is called when the window really gone.
*/
void
gdk_window_destroy_notify (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
GDK_NOTE (EVENTS,
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
g_print ("gdk_window_destroy_notify: %p%s\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
(GDK_WINDOW_DESTROYED (window) ? " (destroyed)" : "")));
if (!GDK_WINDOW_DESTROYED (window))
{
if (GDK_WINDOW_TYPE (window) != GDK_WINDOW_FOREIGN)
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
g_warning ("window %p unexpectedly destroyed",
GDK_WINDOW_HWND (window));
_gdk_window_destroy (window, TRUE);
}
gdk_win32_handle_table_remove (GDK_WINDOW_HWND (window));
g_object_unref (window);
}
static void
get_outer_rect (GdkWindow *window,
gint width,
gint height,
RECT *rect)
{
rect->left = rect->top = 0;
rect->right = width;
rect->bottom = height;
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
_gdk_win32_adjust_client_rect (window, rect);
}
static void
adjust_for_gravity_hints (GdkWindowImplWin32 *impl,
RECT *outer_rect,
gint *x,
gint *y)
{
if (impl->hint_flags & GDK_HINT_WIN_GRAVITY)
{
gint orig_x = *x, orig_y = *y;
switch (impl->hints.win_gravity)
{
case GDK_GRAVITY_NORTH:
case GDK_GRAVITY_CENTER:
case GDK_GRAVITY_SOUTH:
*x -= (outer_rect->right - outer_rect->left) / 2;
*x += impl->width / 2;
break;
case GDK_GRAVITY_SOUTH_EAST:
case GDK_GRAVITY_EAST:
case GDK_GRAVITY_NORTH_EAST:
*x -= outer_rect->right - outer_rect->left;
*x += impl->width;
break;
case GDK_GRAVITY_STATIC:
*x += outer_rect->left;
break;
default:
break;
}
switch (impl->hints.win_gravity)
{
case GDK_GRAVITY_WEST:
case GDK_GRAVITY_CENTER:
case GDK_GRAVITY_EAST:
*y -= (outer_rect->bottom - outer_rect->top) / 2;
*y += impl->height / 2;
break;
case GDK_GRAVITY_SOUTH_WEST:
case GDK_GRAVITY_SOUTH:
case GDK_GRAVITY_SOUTH_EAST:
*y -= outer_rect->bottom - outer_rect->top;
*y += impl->height;
break;
case GDK_GRAVITY_STATIC:
*y += outer_rect->top;
break;
default:
break;
}
GDK_NOTE (MISC,
(orig_x != *x || orig_y != *y) ?
g_print ("adjust_for_gravity_hints: x: %d->%d, y: %d->%d\n",
orig_x, *x, orig_y, *y)
: (void) 0);
}
}
static void
show_window_internal (GdkWindow *window,
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
gboolean raise,
gboolean deiconify)
{
GdkWindowObject *private;
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
HWND old_active_window;
gboolean focus_on_map = TRUE;
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
private = (GdkWindowObject *) window;
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
if (private->destroyed)
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("show_window_internal: %p: %s%s%s\n",
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (private->state),
(raise ? " raise" : ""),
(deiconify ? " deiconify" : "")));
/* If asked to show (not deiconify) an withdrawn and iconified
* window, do that.
*/
if (!deiconify &&
!GDK_WINDOW_IS_MAPPED (window) &&
(private->state & GDK_WINDOW_STATE_ICONIFIED))
{
ShowWindow (GDK_WINDOW_HWND (window), SW_MINIMIZE);
return;
}
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
/* If asked to just show an iconified window, do nothing. */
if (!deiconify && (private->state & GDK_WINDOW_STATE_ICONIFIED))
return;
/* If asked to deiconify an already noniconified window, do
* nothing. (Especially, don't cause the window to rise and
* activate. There are different calls for that.)
*/
if (deiconify && !(private->state & GDK_WINDOW_STATE_ICONIFIED))
return;
/* If asked to show (but not raise) a window that is already
* visible, do nothing.
*/
if (!deiconify && !raise && IsWindowVisible (GDK_WINDOW_HWND (window)))
return;
/* Other cases */
if (!GDK_WINDOW_IS_MAPPED (window))
{
gdk_synthesize_window_state (window,
GDK_WINDOW_STATE_WITHDRAWN,
0);
focus_on_map = private->focus_on_map;
}
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
/* Use SetWindowPos to show transparent windows so automatic redraws
* in other windows can be suppressed.
*/
if (GetWindowLong (GDK_WINDOW_HWND (window), GWL_EXSTYLE) & WS_EX_TRANSPARENT)
{
UINT flags = SWP_SHOWWINDOW | SWP_NOREDRAW | SWP_NOMOVE | SWP_NOSIZE;
if (!raise)
flags |= SWP_NOZORDER;
if (!raise || GDK_WINDOW_TYPE (window) == GDK_WINDOW_TEMP || !focus_on_map)
flags |= SWP_NOACTIVATE;
SetWindowPos (GDK_WINDOW_HWND (window), HWND_TOP, 0, 0, 0, 0, flags);
return;
}
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
old_active_window = GetActiveWindow ();
if (private->state & (GDK_WINDOW_STATE_BELOW | GDK_WINDOW_STATE_ABOVE))
{
DWORD exstyle = GetWindowLong (GDK_WINDOW_HWND (window), GWL_EXSTYLE);
if (private->state & GDK_WINDOW_STATE_BELOW)
exstyle &= (~WS_EX_TOPMOST);
if (private->state & GDK_WINDOW_STATE_ABOVE)
exstyle |= WS_EX_TOPMOST;
API_CALL (SetWindowLong, (GDK_WINDOW_HWND (window), GWL_EXSTYLE, exstyle));
}
if (private->state & GDK_WINDOW_STATE_FULLSCREEN)
gdk_window_fullscreen (window);
else if (private->state & GDK_WINDOW_STATE_MAXIMIZED)
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
ShowWindow (GDK_WINDOW_HWND (window), SW_MAXIMIZE);
else if (private->state & GDK_WINDOW_STATE_ICONIFIED)
ShowWindow (GDK_WINDOW_HWND (window), SW_RESTORE);
else if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_TEMP || !focus_on_map)
Merge from stable: 2003-01-19 Tor Lillqvist <tml@iki.fi> Merge from stable: Fix for #103614 and some other problems with GtkFileSelection on Windows: * gtk/fnmatch.c (get_char): Need to use g_unichar_tolower(), not g_ascii_tolower(). Windows file names are case-insensitive for all Unicode letters. * gtk/gtkfilesel.c: Instead of checking for G_OS_WIN32 or G_WITH_CYGWIN, check G_PLATFORM_WIN32. Move inclusion of gtkintl.h earlier, as it includes config.h unconditionally, and gtkprivate.h redefines GTK_LOCALEDIR. (struct _CompletionDirSent): Ifdef out the fields not used on Windows. (compare_utf8_filenames, compare_sys_filenames): Need different comparison implementation for UTF-8 file names and system locale file names on Win32. Cannot simply use g_ascii_strcasecmp(), but need to casefold all Unicode letters. (cmpl_completion_matches, open_dir, correct_parent): Ifdef out variables not used on Win32 to avoid warnings about unused variables. (open_ref_dir): Use g_path_skip_root() to skip past potential drive letter in front of the leading (back)slash. (open_new_dir): Ifdef out use of CompletionDirSent fields not there on Win32. (correct_parent): Bypass inode check also on Cygwin. Fix bug noticed by Alex Shaduri: Tooltips and other GDK_WINDOW_TEMP windows were activated. This looked very odd, and was a regression from earlier versions. * gdk/win32/gdkwindow-win32.c (show_window_internal): Fine-tune behaviour. Don't ever activate GDK_WINDOW_TEMP windows. (gdk_window_move, gdk_window_resize): Add debug logging. * gdk/win32/gdkevents-win32.c (gdk_event_translate): When we get a WM_SIZE message for a non-visible (withdrawn) window, don't clear the GDK_WINDOW_STATE_WITHDRAWN bit. The window is still withdrawn even if its size changes.
2003-01-19 21:18:15 +00:00
ShowWindow (GDK_WINDOW_HWND (window), SW_SHOWNOACTIVATE);
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
else
ShowWindow (GDK_WINDOW_HWND (window), SW_SHOWNORMAL);
if (raise)
{
if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_TEMP)
SetWindowPos (GDK_WINDOW_HWND (window), HWND_TOPMOST,
0, 0, 0, 0,
SWP_NOACTIVATE | SWP_NOMOVE | SWP_NOSIZE);
else if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_TOPLEVEL ||
GDK_WINDOW_TYPE (window) == GDK_WINDOW_DIALOG)
{
if (focus_on_map && private->accept_focus)
SetForegroundWindow (GDK_WINDOW_HWND (window));
else
SetWindowPos (GDK_WINDOW_HWND (window), HWND_TOP,
0, 0, 0, 0,
SWP_NOACTIVATE | SWP_NOMOVE | SWP_NOSIZE);
}
else
BringWindowToTop (GDK_WINDOW_HWND (window));
}
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
else if (old_active_window != GDK_WINDOW_HWND (window))
SetActiveWindow (old_active_window);
}
void
gdk_window_show_unraised (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
show_window_internal (window, FALSE, FALSE);
}
void
gdk_window_show (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
show_window_internal (window, TRUE, FALSE);
}
void
gdk_window_hide (GdkWindow *window)
{
GdkWindowObject *private;
g_return_if_fail (GDK_IS_WINDOW (window));
private = (GdkWindowObject*) window;
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
if (private->destroyed)
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_hide: %p: %s\n",
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (private->state)));
if (GDK_WINDOW_IS_MAPPED (window))
gdk_synthesize_window_state (window,
0,
GDK_WINDOW_STATE_WITHDRAWN);
_gdk_window_clear_update_area (window);
if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_TOPLEVEL)
ShowOwnedPopups (GDK_WINDOW_HWND (window), FALSE);
if (GetWindowLong (GDK_WINDOW_HWND (window), GWL_EXSTYLE) & WS_EX_TRANSPARENT)
{
SetWindowPos (GDK_WINDOW_HWND (window), HWND_BOTTOM,
0, 0, 0, 0,
SWP_HIDEWINDOW | SWP_NOREDRAW | SWP_NOZORDER | SWP_NOMOVE | SWP_NOSIZE);
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
}
else
{
ShowWindow (GDK_WINDOW_HWND (window), SW_HIDE);
}
}
void
gdk_window_withdraw (GdkWindow *window)
{
GdkWindowObject *private;
g_return_if_fail (GDK_IS_WINDOW (window));
private = (GdkWindowObject*) window;
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
if (private->destroyed)
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_withdraw: %p: %s\n",
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (private->state)));
gdk_window_hide (window); /* ??? */
}
void
gdk_window_move (GdkWindow *window,
gint x,
gint y)
{
GdkWindowObject *private = (GdkWindowObject *)window;
GdkWindowImplWin32 *impl;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
GDK_NOTE (MISC, g_print ("gdk_window_move: %p: %+d%+d\n",
Merge from stable: 2003-01-19 Tor Lillqvist <tml@iki.fi> Merge from stable: Fix for #103614 and some other problems with GtkFileSelection on Windows: * gtk/fnmatch.c (get_char): Need to use g_unichar_tolower(), not g_ascii_tolower(). Windows file names are case-insensitive for all Unicode letters. * gtk/gtkfilesel.c: Instead of checking for G_OS_WIN32 or G_WITH_CYGWIN, check G_PLATFORM_WIN32. Move inclusion of gtkintl.h earlier, as it includes config.h unconditionally, and gtkprivate.h redefines GTK_LOCALEDIR. (struct _CompletionDirSent): Ifdef out the fields not used on Windows. (compare_utf8_filenames, compare_sys_filenames): Need different comparison implementation for UTF-8 file names and system locale file names on Win32. Cannot simply use g_ascii_strcasecmp(), but need to casefold all Unicode letters. (cmpl_completion_matches, open_dir, correct_parent): Ifdef out variables not used on Win32 to avoid warnings about unused variables. (open_ref_dir): Use g_path_skip_root() to skip past potential drive letter in front of the leading (back)slash. (open_new_dir): Ifdef out use of CompletionDirSent fields not there on Win32. (correct_parent): Bypass inode check also on Cygwin. Fix bug noticed by Alex Shaduri: Tooltips and other GDK_WINDOW_TEMP windows were activated. This looked very odd, and was a regression from earlier versions. * gdk/win32/gdkwindow-win32.c (show_window_internal): Fine-tune behaviour. Don't ever activate GDK_WINDOW_TEMP windows. (gdk_window_move, gdk_window_resize): Add debug logging. * gdk/win32/gdkevents-win32.c (gdk_event_translate): When we get a WM_SIZE message for a non-visible (withdrawn) window, don't clear the GDK_WINDOW_STATE_WITHDRAWN bit. The window is still withdrawn even if its size changes.
2003-01-19 21:18:15 +00:00
GDK_WINDOW_HWND (window), x, y));
impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
if (private->state & GDK_WINDOW_STATE_FULLSCREEN)
return;
/* Don't check GDK_WINDOW_TYPE (private) == GDK_WINDOW_CHILD.
* Foreign windows (another app's windows) might be children of our
* windows! Especially in the case of gtkplug/socket.
*/
if (GetAncestor (GDK_WINDOW_HWND (window), GA_PARENT) != GetDesktopWindow ())
{
_gdk_window_move_resize_child (window, x, y, impl->width, impl->height);
}
else
{
RECT outer_rect;
get_outer_rect (window, impl->width, impl->height, &outer_rect);
adjust_for_gravity_hints (impl, &outer_rect, &x, &y);
GDK_NOTE (MISC, g_print ("... SetWindowPos(%p,NULL,%d,%d,0,0,"
"NOACTIVATE|NOSIZE|NOZORDER)\n",
GDK_WINDOW_HWND (window),
x - _gdk_offset_x, y - _gdk_offset_y));
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), NULL,
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
x - _gdk_offset_x, y - _gdk_offset_y, 0, 0,
SWP_NOACTIVATE | SWP_NOSIZE | SWP_NOZORDER));
}
}
void
gdk_window_resize (GdkWindow *window,
gint width,
gint height)
{
GdkWindowObject *private = (GdkWindowObject*) window;
GdkWindowImplWin32 *impl;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
if (width < 1)
width = 1;
if (height < 1)
height = 1;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_resize: %p: %dx%d\n",
Merge from stable: 2003-01-19 Tor Lillqvist <tml@iki.fi> Merge from stable: Fix for #103614 and some other problems with GtkFileSelection on Windows: * gtk/fnmatch.c (get_char): Need to use g_unichar_tolower(), not g_ascii_tolower(). Windows file names are case-insensitive for all Unicode letters. * gtk/gtkfilesel.c: Instead of checking for G_OS_WIN32 or G_WITH_CYGWIN, check G_PLATFORM_WIN32. Move inclusion of gtkintl.h earlier, as it includes config.h unconditionally, and gtkprivate.h redefines GTK_LOCALEDIR. (struct _CompletionDirSent): Ifdef out the fields not used on Windows. (compare_utf8_filenames, compare_sys_filenames): Need different comparison implementation for UTF-8 file names and system locale file names on Win32. Cannot simply use g_ascii_strcasecmp(), but need to casefold all Unicode letters. (cmpl_completion_matches, open_dir, correct_parent): Ifdef out variables not used on Win32 to avoid warnings about unused variables. (open_ref_dir): Use g_path_skip_root() to skip past potential drive letter in front of the leading (back)slash. (open_new_dir): Ifdef out use of CompletionDirSent fields not there on Win32. (correct_parent): Bypass inode check also on Cygwin. Fix bug noticed by Alex Shaduri: Tooltips and other GDK_WINDOW_TEMP windows were activated. This looked very odd, and was a regression from earlier versions. * gdk/win32/gdkwindow-win32.c (show_window_internal): Fine-tune behaviour. Don't ever activate GDK_WINDOW_TEMP windows. (gdk_window_move, gdk_window_resize): Add debug logging. * gdk/win32/gdkevents-win32.c (gdk_event_translate): When we get a WM_SIZE message for a non-visible (withdrawn) window, don't clear the GDK_WINDOW_STATE_WITHDRAWN bit. The window is still withdrawn even if its size changes.
2003-01-19 21:18:15 +00:00
GDK_WINDOW_HWND (window), width, height));
impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
if (private->state & GDK_WINDOW_STATE_FULLSCREEN)
return;
if (GetAncestor (GDK_WINDOW_HWND (window), GA_PARENT) != GetDesktopWindow ())
{
_gdk_window_move_resize_child (window, private->x, private->y, width, height);
}
else
{
RECT outer_rect;
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
get_outer_rect (window, width, height, &outer_rect);
GDK_NOTE (MISC, g_print ("... SetWindowPos(%p,NULL,0,0,%ld,%ld,"
"NOACTIVATE|NOMOVE|NOZORDER)\n",
GDK_WINDOW_HWND (window),
outer_rect.right - outer_rect.left,
outer_rect.bottom - outer_rect.top));
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), NULL,
0, 0,
outer_rect.right - outer_rect.left,
outer_rect.bottom - outer_rect.top,
SWP_NOACTIVATE | SWP_NOMOVE | SWP_NOZORDER));
private->resize_count += 1;
}
}
void
gdk_window_move_resize (GdkWindow *window,
gint x,
gint y,
gint width,
gint height)
{
GdkWindowObject *private = (GdkWindowObject*) window;
GdkWindowImplWin32 *impl;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
if (width < 1)
width = 1;
if (height < 1)
height = 1;
impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
if (private->state & GDK_WINDOW_STATE_FULLSCREEN)
return;
GDK_NOTE (MISC, g_print ("gdk_window_move_resize: %p: %dx%d@%+d%+d\n",
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
GDK_WINDOW_HWND (window),
width, height, x, y));
if (GetAncestor (GDK_WINDOW_HWND (window), GA_PARENT) != GetDesktopWindow ())
{
_gdk_window_move_resize_child (window, x, y, width, height);
}
else
gdk/win32/gdkprivate-win32.h New flags _gdk_input_locale_is_ime and 2003-07-25 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New flags _gdk_input_locale_is_ime and _gdk_keyboard_has_altgr. * gdk/win32/gdkevents-win32.c: Lots of changes. Most important ones detailled here. Code that has been ifdeffed out for a long time removed. Remove some really old doc comments that were left behind for some public functions, the official ones are in the X11 backend anyway. Change GDK_WINDOW_OBJECT() calls to GdkWindowObject casts. Reformat multi-line boolean expressions to have the operators at ends of lines. As mouse capture with SetCapture() indeed seems to work OK, no need to have the correspoinding macro USE_SETCAPTURE and ifdefs. Ifdef out the gdk-ping-msg stuff. I don't remember why it was needed at some time, and things seem to work fine now without (knock on wood). Ifdef out the search for some Latin locale keyboard layout being loaded. Not used currently, but might be needed after all, if we decide that we want to be able to generate ASCII control character events with a non-Latin keyboard. (assign_object): New helper function, handles the g_object_ref() and unref() calls when assigning GObject pointers. (generate_crossing_events): Also generate the GDK_NOTIFY_INTERIOR enter event when the pointer has moved to an ancestor window. Was left out by mistake. (gdk_window_is_ancestor): Renamed from gdk_window_is_child(). (gdk_pointer_grab, gdk_pointer_ungrab): Implement the confine_to functionality, using ClipCursor(). (find_window_for_mouse_event): Splice part of code into new function find_real_window_for_grabbed_mouse_event(). (fixup_event, append_event, apply_filters): New functions, code refactored out from elsewhere. (synthesize_enter_or_leave_event, synthesize_leave_event, synthesize_enter_event, synthesize_leave_events,synthesize_enter_events): Also take a GdkCrossingMode parameter, in preparation to generating GDK_CROSSING_GRAB and GDK_CROSSING_UNGRAB events. (fixup_event, append_event, fill_key_event_string): New functions, code refactoring. (vk_from_char, build_keypress_event, build_keyrelease_event): Removed as part of dropping WM_CHAR handling. (build_key_event_state,gdk_event_translate): Call GetKeyboardState(), once, for each keyboard message, instead of several calls to GetKeyState() here and there. (gdk_event_translate): Fix bugs #104516, #104662, #115902. While at it, do some major refactoring, and some fixes for potential problems noticed while going through the code. Don't handle WM_CHAR at all. Only handle WM_KEYDOWN and WM_KEYUP. Don't need the state variables related to whether to wait for WM_CHAR or not, and whether the current key is AltGr. Remove lots of complexity. Thus don't need the use_ime_composition flag. Not handling WM_CHAR means dead key handling will have to be taken care of by GTK, but that seems to work fine, so no worry. Another side-effect is that Alt+keypad digits don't work any longer, but it's better to learn to use GTK's ISO14755 support is anyway. Be more careful in checking whether AltGr is involved. Only attempt to handle it if the keyboard actually has it. And explicitly check for *left* Control plus *right* Alt being pressed. Still, allow (left) Alt and/or (right) Control with AltGr chars. Handle keys using similar code as in the X11 backend. As we have built a keymap in gdkkeys-win32.c anyway, use it by calling gdk_keymap_translate_keyboard_state() to look up the keysym from the virtual key code and keyboard state. Build the key event string in exactly the same way as the X11 backend. If an IME is being used, don't generate GDK events for keys between receiving WM_IME_STARTCOMPOSITION and WM_IME_ENDCOMPOSITION, as those keys are for the IME. For WM_IME_COMPOSITION, handle all the Unicode chars returned from the IME, not just the first one. gdk_event_translate() is still quite complex, could split the message handler cases out into separate functions. On mouse events, when the mouse is grabbed, use find_real_window_for_grabbed_mouse_event() in order to be able to generate correct crossing events. No longer take a pre-allocated GdkEvent as parameter. Instead, allocate events as needed and append them to the queue. (This is different from how gdk_event_translate() in the X11 backend works.) This change made the code much clearer, especially in the cases where we have to generate several GDK events for one Windows message. Return FALSE if DefWindowProc() should be called, TRUE if not. If DefWindowProc() should not be called, also return the value to be returned from the window procedure. (Previously, the interaction with gdk_event_translate()'s caller was much more complex, when we had to indicate whether the already-queued event should be left in the queue or removed, and in addition also had to indicate whether to call DefWindowProc() or not, and what value to return from the window procedure if not.) Don't use a separate "private" variable required to be pointing to the GdkWindowObject of the "window" variable at all times. Just use casts, even if looks a bit uglier. Notice destroyed windows as early as possible, and break out of the messsage switch. Use _gdk_pointer_root as current_window when the pointer is outside GDK's top-level windows. On WM_INPUTLANGCHANGE, set _gdk_input_locale_is_ime as appropriate, based on ImmIsIME(). (gdk_event_translate, gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Implement client messages. Use a registered Windows message to pass GDK client messages. Note that the amount of user data is restricted to four bytes, as it is carried in the LPARAM. (The WPARAM is used for the message type "atom".) (real_window_procedure): Adapt for new gdk_event_translate() interface. * gdk/win32/gdkmain-win32.c (_gdk_windowing_init): Set _gdk_input_locale_is_ime initially. * gdk/win32/gdkwindow-win32.c: Use g_object_ref()/unref() instead of g_colormap_ref()/unref(). (gdk_window_new): Made code a bit more like the X11 one, pretend to handle screens (although we just have one for now). * gdk/x11/gdkevents-x11.c (gdk_event_send_client_message_for_display, gdk_screen_broadcast_client_message): Document the user data limitation on Win32. * gdk/win32/gdkevents-win32.c (print_event): More complete enter and leave notify detail output. * gdk/win32/gdkkeys-win32.c (update_keymap): Make dead keys visible to GDK and GTK. Store the corresponding GDK_dead_* keysym for those, so that the GtkIMContextCimple compose tables will work. Deduce if the keyboard layout has the AltGr key, and set the above flag accordingly.
2003-07-26 01:54:59 +00:00
{
RECT outer_rect;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
get_outer_rect (window, width, height, &outer_rect);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
adjust_for_gravity_hints (impl, &outer_rect, &x, &y);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("... SetWindowPos(%p,NULL,%d,%d,%ld,%ld,"
"NOACTIVATE|NOZORDER)\n",
GDK_WINDOW_HWND (window),
x - _gdk_offset_x, y - _gdk_offset_y,
outer_rect.right - outer_rect.left,
outer_rect.bottom - outer_rect.top));
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), NULL,
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
x - _gdk_offset_x, y - _gdk_offset_y,
outer_rect.right - outer_rect.left,
outer_rect.bottom - outer_rect.top,
SWP_NOACTIVATE | SWP_NOZORDER));
}
}
void
gdk_window_reparent (GdkWindow *window,
GdkWindow *new_parent,
gint x,
gint y)
{
GdkWindowObject *window_private;
GdkWindowObject *parent_private;
GdkWindowObject *old_parent_private;
GdkWindowImplWin32 *impl;
gboolean was_toplevel;
LONG style;
g_return_if_fail (GDK_IS_WINDOW (window));
g_return_if_fail (new_parent == NULL || GDK_IS_WINDOW (new_parent));
g_return_if_fail (window != _gdk_root);
if (GDK_WINDOW_DESTROYED (window) ||
(new_parent && GDK_WINDOW_DESTROYED (new_parent)))
{
return;
}
if (!new_parent)
new_parent = _gdk_root;
window_private = (GdkWindowObject*) window;
old_parent_private = (GdkWindowObject *) window_private->parent;
parent_private = (GdkWindowObject*) new_parent;
impl = GDK_WINDOW_IMPL_WIN32 (window_private->impl);
GDK_NOTE (MISC, g_print ("gdk_window_reparent: %p: %p\n",
GDK_WINDOW_HWND (window),
GDK_WINDOW_HWND (new_parent)));
style = GetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
was_toplevel = GetAncestor (GDK_WINDOW_HWND (window), GA_PARENT) == GetDesktopWindow ();
if (was_toplevel && new_parent != _gdk_root)
{
/* Reparenting from top-level (child of desktop). Clear out
* decorations.
*/
style &= ~(WS_OVERLAPPED | WS_CAPTION | WS_SYSMENU | WS_THICKFRAME | WS_MINIMIZEBOX | WS_MAXIMIZEBOX);
style |= WS_CHILD;
SetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE, style);
}
else if (new_parent == _gdk_root)
{
/* Reparenting to top-level. Add decorations. */
style &= ~(WS_CHILD);
style |= WS_OVERLAPPEDWINDOW;
SetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE, style);
}
API_CALL (SetParent, (GDK_WINDOW_HWND (window),
GDK_WINDOW_HWND (new_parent)));
API_CALL (MoveWindow, (GDK_WINDOW_HWND (window),
x, y, impl->width, impl->height, TRUE));
/* From here on, we treat parents of type GDK_WINDOW_FOREIGN like
* the root window
*/
if (GDK_WINDOW_TYPE (new_parent) == GDK_WINDOW_FOREIGN)
new_parent = _gdk_root;
window_private->parent = (GdkWindowObject *)new_parent;
/* Switch the window type as appropriate */
switch (GDK_WINDOW_TYPE (new_parent))
{
case GDK_WINDOW_ROOT:
if (impl->toplevel_window_type != -1)
GDK_WINDOW_TYPE (window) = impl->toplevel_window_type;
else if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_CHILD)
GDK_WINDOW_TYPE (window) = GDK_WINDOW_TOPLEVEL;
break;
case GDK_WINDOW_TOPLEVEL:
case GDK_WINDOW_CHILD:
case GDK_WINDOW_DIALOG:
case GDK_WINDOW_TEMP:
if (WINDOW_IS_TOPLEVEL (window))
{
/* Save the original window type so we can restore it if the
* window is reparented back to be a toplevel.
*/
impl->toplevel_window_type = GDK_WINDOW_TYPE (window);
GDK_WINDOW_TYPE (window) = GDK_WINDOW_CHILD;
}
}
if (old_parent_private)
old_parent_private->children =
g_list_remove (old_parent_private->children, window);
parent_private->children = g_list_prepend (parent_private->children, window);
_gdk_window_init_position (GDK_WINDOW (window_private));
}
static void
erase_background (GdkWindow *window,
HDC hdc)
{
HDC bgdc = NULL;
HBRUSH hbr = NULL;
HPALETTE holdpal = NULL;
RECT rect;
COLORREF bg;
GdkColormap *colormap;
GdkColormapPrivateWin32 *colormap_private;
int x, y;
int x_offset, y_offset;
if (((GdkWindowObject *) window)->input_only ||
((GdkWindowObject *) window)->bg_pixmap == GDK_NO_BG ||
GDK_WINDOW_IMPL_WIN32 (((GdkWindowObject *) window)->impl)->position_info.no_bg)
{
return;
}
colormap = gdk_drawable_get_colormap (window);
if (colormap &&
(colormap->visual->type == GDK_VISUAL_PSEUDO_COLOR ||
colormap->visual->type == GDK_VISUAL_STATIC_COLOR))
{
int k;
colormap_private = GDK_WIN32_COLORMAP_DATA (colormap);
if (!(holdpal = SelectPalette (hdc, colormap_private->hpal, FALSE)))
WIN32_GDI_FAILED ("SelectPalette");
else if ((k = RealizePalette (hdc)) == GDI_ERROR)
WIN32_GDI_FAILED ("RealizePalette");
else if (k > 0)
GDK_NOTE (COLORMAP, g_print ("erase_background: realized %p: %d colors\n",
colormap_private->hpal, k));
}
x_offset = y_offset = 0;
while (window && ((GdkWindowObject *) window)->bg_pixmap == GDK_PARENT_RELATIVE_BG)
{
/* If this window should have the same background as the parent,
* fetch the parent. (And if the same goes for the parent, fetch
* the grandparent, etc.)
*/
x_offset += ((GdkWindowObject *) window)->x;
y_offset += ((GdkWindowObject *) window)->y;
window = GDK_WINDOW (((GdkWindowObject *) window)->parent);
}
if (GDK_WINDOW_IMPL_WIN32 (((GdkWindowObject *) window)->impl)->position_info.no_bg)
{
/* Improves scolling effect, e.g. main buttons of testgtk */
return;
}
GetClipBox (hdc, &rect);
if (((GdkWindowObject *) window)->bg_pixmap == NULL)
{
bg = _gdk_win32_colormap_color (GDK_DRAWABLE_IMPL_WIN32 (((GdkWindowObject *) window)->impl)->colormap,
((GdkWindowObject *) window)->bg_color.pixel);
if (!(hbr = CreateSolidBrush (bg)))
WIN32_GDI_FAILED ("CreateSolidBrush");
else if (!FillRect (hdc, &rect, hbr))
WIN32_GDI_FAILED ("FillRect");
if (hbr != NULL)
DeleteObject (hbr);
}
else if (((GdkWindowObject *) window)->bg_pixmap != GDK_NO_BG)
{
GdkPixmap *pixmap = ((GdkWindowObject *) window)->bg_pixmap;
GdkPixmapImplWin32 *pixmap_impl = GDK_PIXMAP_IMPL_WIN32 (GDK_PIXMAP_OBJECT (pixmap)->impl);
if (x_offset == 0 && y_offset == 0 &&
pixmap_impl->width <= 8 && pixmap_impl->height <= 8)
{
if (!(hbr = CreatePatternBrush (GDK_PIXMAP_HBITMAP (pixmap))))
WIN32_GDI_FAILED ("CreatePatternBrush");
else if (!FillRect (hdc, &rect, hbr))
WIN32_GDI_FAILED ("FillRect");
if (hbr != NULL)
DeleteObject (hbr);
}
else
{
HGDIOBJ oldbitmap;
if (!(bgdc = CreateCompatibleDC (hdc)))
{
WIN32_GDI_FAILED ("CreateCompatibleDC");
return;
}
if (!(oldbitmap = SelectObject (bgdc, GDK_PIXMAP_HBITMAP (pixmap))))
{
WIN32_GDI_FAILED ("SelectObject");
DeleteDC (bgdc);
return;
}
x = -x_offset;
while (x < rect.right)
{
if (x + pixmap_impl->width >= rect.left)
{
y = -y_offset;
while (y < rect.bottom)
{
if (y + pixmap_impl->height >= rect.top)
{
if (!BitBlt (hdc, x, y,
pixmap_impl->width, pixmap_impl->height,
bgdc, 0, 0, SRCCOPY))
{
WIN32_GDI_FAILED ("BitBlt");
SelectObject (bgdc, oldbitmap);
DeleteDC (bgdc);
return;
}
}
y += pixmap_impl->height;
}
}
x += pixmap_impl->width;
}
SelectObject (bgdc, oldbitmap);
DeleteDC (bgdc);
}
}
}
void
Large changes to the Win32 backend, partially made necessary by the 2000-05-02 Tor Lillqvist <tml@iki.fi> Large changes to the Win32 backend, partially made necessary by the changes to the backend-independent internal structures. Attempts to implement similar backing store stuff as on X11. The current (CVS) version of the Win32 backend is *not* as stable as it was before the no-flicker branch was merged. A zipfile with that version is available from http://www.gimp.org/win32/. That should be use by "production" code until this CVS version is usable. (But note, the Win32 backend has never been claimed to be "production quality".) * README.win32: Add the above comment about versions. * gdk/gdkwindow.c: Don't use backing store for now on Win32. * gdk/gdk.def: Update. * gdk/gdkfont.h: Declare temporary Win32-only functions. Will presumably be replaced by some more better mechanism as 1.4 gets closer to release shape. * gdk/makefile.{cygwin,msc}: Update. * gdk/win32/*.c: Correct inclusions of the backend-specific and internal headers. Change code according to changes in these. Use gdk_drawable_*, not gdk_window_* where necessary. * gdk/win32/gdkdnd-win32.c: Use MISC selector for GDK_NOTE, not our old DND. * gdk/win32/gdkdrawable-win32.c (gdk_win32_draw_text): Don't try to interpret single characters as UTF-8. Thanks to Hans Breuer. Use correct function name in warning messages. * gdk/win32/gdkevents-win32.c: Use correct parameter lists for the GSourceFuncs gdk_event_prepare and gdk_event_check. (gdk_event_get_graphics_expose): Do implement, use PeekMessage. Thanks to Hans Breuer. (event_mask_string): Debugging function to print an GdkEventMask. (gdk_pointer_grab): Use it. * gdk/win32/gdkfont-win32.c: The Unicode subrange that the (old) book I used claimed was Hangul actually is CJK Unified Ideographs Extension A. Also, Hangul Syllables were missing. Improve logging. * gdk/win32/gdkgc-win32.c: Largish changes. * gdk/win32/gdkim-win32.c (gdk_set_locale): Use g_win32_getlocale() from GLib, and not setlocale() to get current locale name. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwin32.h: Move stuff from gdkprivate-win32.h to gdkwin32.h, similarily as in the X11 backend. * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Bugfix, assignment was used instead of equals in if test. Thanks to Hans Breuer. * gdk/win32/makefile.{cygwin,msc} * gtk/makefile.{cygwin,msc}: Updates. Better kludge to get the path to the Win32 headers that works also with the mingw compiler. * gtk/gtkstyle.c: Include <string.h>.
2000-05-01 22:06:49 +00:00
_gdk_windowing_window_clear_area (GdkWindow *window,
gint x,
gint y,
gint width,
gint height)
{
GdkWindowImplWin32 *impl;
g_return_if_fail (GDK_IS_WINDOW (window));
impl = GDK_WINDOW_IMPL_WIN32 (GDK_WINDOW_OBJECT (window)->impl);
if (!GDK_WINDOW_DESTROYED (window))
{
HDC hdc;
if (width == 0)
width = impl->width - x;
if (height == 0)
height = impl->height - y;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("_gdk_windowing_window_clear_area: %p: "
"%dx%d@%+d%+d\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
width, height, x, y));
hdc = GetDC (GDK_WINDOW_HWND (window));
IntersectClipRect (hdc, x, y, x + width, y + height);
erase_background (window, hdc);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDI_CALL (ReleaseDC, (GDK_WINDOW_HWND (window), hdc));
}
}
void
Large changes to the Win32 backend, partially made necessary by the 2000-05-02 Tor Lillqvist <tml@iki.fi> Large changes to the Win32 backend, partially made necessary by the changes to the backend-independent internal structures. Attempts to implement similar backing store stuff as on X11. The current (CVS) version of the Win32 backend is *not* as stable as it was before the no-flicker branch was merged. A zipfile with that version is available from http://www.gimp.org/win32/. That should be use by "production" code until this CVS version is usable. (But note, the Win32 backend has never been claimed to be "production quality".) * README.win32: Add the above comment about versions. * gdk/gdkwindow.c: Don't use backing store for now on Win32. * gdk/gdk.def: Update. * gdk/gdkfont.h: Declare temporary Win32-only functions. Will presumably be replaced by some more better mechanism as 1.4 gets closer to release shape. * gdk/makefile.{cygwin,msc}: Update. * gdk/win32/*.c: Correct inclusions of the backend-specific and internal headers. Change code according to changes in these. Use gdk_drawable_*, not gdk_window_* where necessary. * gdk/win32/gdkdnd-win32.c: Use MISC selector for GDK_NOTE, not our old DND. * gdk/win32/gdkdrawable-win32.c (gdk_win32_draw_text): Don't try to interpret single characters as UTF-8. Thanks to Hans Breuer. Use correct function name in warning messages. * gdk/win32/gdkevents-win32.c: Use correct parameter lists for the GSourceFuncs gdk_event_prepare and gdk_event_check. (gdk_event_get_graphics_expose): Do implement, use PeekMessage. Thanks to Hans Breuer. (event_mask_string): Debugging function to print an GdkEventMask. (gdk_pointer_grab): Use it. * gdk/win32/gdkfont-win32.c: The Unicode subrange that the (old) book I used claimed was Hangul actually is CJK Unified Ideographs Extension A. Also, Hangul Syllables were missing. Improve logging. * gdk/win32/gdkgc-win32.c: Largish changes. * gdk/win32/gdkim-win32.c (gdk_set_locale): Use g_win32_getlocale() from GLib, and not setlocale() to get current locale name. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwin32.h: Move stuff from gdkprivate-win32.h to gdkwin32.h, similarily as in the X11 backend. * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Bugfix, assignment was used instead of equals in if test. Thanks to Hans Breuer. * gdk/win32/makefile.{cygwin,msc} * gtk/makefile.{cygwin,msc}: Updates. Better kludge to get the path to the Win32 headers that works also with the mingw compiler. * gtk/gtkstyle.c: Include <string.h>.
2000-05-01 22:06:49 +00:00
_gdk_windowing_window_clear_area_e (GdkWindow *window,
gint x,
gint y,
gint width,
gint height)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (!GDK_WINDOW_DESTROYED (window))
{
HDC hdc;
RECT rect;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("_gdk_windowing_window_clear_area_e: %p: "
"%dx%d@%+d%+d\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
width, height, x, y));
/* The background should be erased before the expose event is
generated */
hdc = GetDC (GDK_WINDOW_HWND (window));
IntersectClipRect (hdc, x, y, x + width, y + height);
erase_background (window, hdc);
GDI_CALL (ReleaseDC, (GDK_WINDOW_HWND (window), hdc));
rect.left = x;
rect.right = x + width;
rect.top = y;
rect.bottom = y + height;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDI_CALL (InvalidateRect, (GDK_WINDOW_HWND (window), &rect, TRUE));
UpdateWindow (GDK_WINDOW_HWND (window));
}
}
void
gdk_window_raise (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (!GDK_WINDOW_DESTROYED (window))
{
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_raise: %p\n",
GDK_WINDOW_HWND (window)));
if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_TEMP)
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), HWND_TOPMOST,
0, 0, 0, 0,
SWP_NOACTIVATE | SWP_NOMOVE | SWP_NOSIZE));
else if (((GdkWindowObject *)window)->accept_focus)
API_CALL (BringWindowToTop, (GDK_WINDOW_HWND (window)));
else
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), HWND_TOP,
0, 0, 0, 0,
SWP_NOACTIVATE | SWP_NOMOVE | SWP_NOSIZE));
}
}
void
gdk_window_lower (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (!GDK_WINDOW_DESTROYED (window))
{
GDK_NOTE (MISC, g_print ("gdk_window_lower: %p\n"
"... SetWindowPos(%p,HWND_BOTTOM,0,0,0,0,"
"NOACTIVATE|NOMOVE|NOSIZE)\n",
GDK_WINDOW_HWND (window),
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window)));
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), HWND_BOTTOM,
0, 0, 0, 0,
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
SWP_NOACTIVATE | SWP_NOMOVE | SWP_NOSIZE));
}
}
void
gdk_window_set_hints (GdkWindow *window,
gint x,
gint y,
gint min_width,
gint min_height,
gint max_width,
gint max_height,
gint flags)
{
/* Note that this function is obsolete */
GdkWindowImplWin32 *impl;
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
impl = GDK_WINDOW_IMPL_WIN32 (GDK_WINDOW_OBJECT (window)->impl);
GDK_NOTE (MISC, g_print ("gdk_window_set_hints: %p: %dx%d..%dx%d @%+d%+d\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
min_width, min_height, max_width, max_height,
x, y));
if (flags)
{
GdkGeometry geom;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
gint geom_mask = 0;
geom.min_width = min_width;
geom.min_height = min_height;
geom.max_width = max_width;
geom.max_height = max_height;
if (flags & GDK_HINT_MIN_SIZE)
geom_mask |= GDK_HINT_MIN_SIZE;
if (flags & GDK_HINT_MAX_SIZE)
geom_mask |= GDK_HINT_MAX_SIZE;
gdk_window_set_geometry_hints (window, &geom, geom_mask);
}
}
void
gdk_window_set_urgency_hint (GdkWindow *window,
gboolean urgent)
{
FLASHWINFO flashwinfo;
typedef BOOL (*PFN_FlashWindowEx) (FLASHWINFO*);
PFN_FlashWindowEx flashWindowEx = NULL;
g_return_if_fail (GDK_IS_WINDOW (window));
g_return_if_fail (GDK_WINDOW_TYPE (window) != GDK_WINDOW_CHILD);
if (GDK_WINDOW_DESTROYED (window))
return;
flashWindowEx = (PFN_FlashWindowEx) GetProcAddress (GetModuleHandle ("user32.dll"), "FlashWindowEx");
if (flashWindowEx)
{
flashwinfo.cbSize = sizeof (flashwinfo);
flashwinfo.hwnd = GDK_WINDOW_HWND (window);
if (urgent)
flashwinfo.dwFlags = FLASHW_ALL | FLASHW_TIMER;
else
flashwinfo.dwFlags = FLASHW_STOP;
flashwinfo.uCount = 0;
flashwinfo.dwTimeout = 0;
flashWindowEx (&flashwinfo);
}
else
{
FlashWindow (GDK_WINDOW_HWND (window), urgent);
}
}
static gboolean
get_effective_window_decorations (GdkWindow *window,
GdkWMDecoration *decoration)
{
GdkWindowImplWin32 *impl;
impl = (GdkWindowImplWin32 *)((GdkWindowObject *)window)->impl;
if (gdk_window_get_decorations (window, decoration))
return TRUE;
if (((GdkWindowObject *) window)->window_type != GDK_WINDOW_TOPLEVEL &&
((GdkWindowObject *) window)->window_type != GDK_WINDOW_DIALOG)
return FALSE;
if ((impl->hint_flags & GDK_HINT_MIN_SIZE) &&
(impl->hint_flags & GDK_HINT_MAX_SIZE) &&
impl->hints.min_width == impl->hints.max_width &&
impl->hints.min_height == impl->hints.max_height)
{
*decoration = GDK_DECOR_ALL | GDK_DECOR_RESIZEH | GDK_DECOR_MAXIMIZE;
if (impl->type_hint == GDK_WINDOW_TYPE_HINT_DIALOG ||
impl->type_hint == GDK_WINDOW_TYPE_HINT_MENU ||
impl->type_hint == GDK_WINDOW_TYPE_HINT_TOOLBAR)
*decoration |= GDK_DECOR_MINIMIZE;
else if (impl->type_hint == GDK_WINDOW_TYPE_HINT_SPLASHSCREEN)
*decoration |= GDK_DECOR_MENU | GDK_DECOR_MINIMIZE;
return TRUE;
}
else if (impl->hint_flags & GDK_HINT_MAX_SIZE)
{
*decoration = GDK_DECOR_ALL | GDK_DECOR_MAXIMIZE;
if (impl->type_hint == GDK_WINDOW_TYPE_HINT_DIALOG ||
impl->type_hint == GDK_WINDOW_TYPE_HINT_MENU ||
impl->type_hint == GDK_WINDOW_TYPE_HINT_TOOLBAR)
*decoration |= GDK_DECOR_MINIMIZE;
return TRUE;
}
else
{
switch (impl->type_hint)
{
case GDK_WINDOW_TYPE_HINT_DIALOG:
*decoration = (GDK_DECOR_ALL | GDK_DECOR_MINIMIZE | GDK_DECOR_MAXIMIZE);
return TRUE;
case GDK_WINDOW_TYPE_HINT_MENU:
*decoration = (GDK_DECOR_ALL | GDK_DECOR_RESIZEH | GDK_DECOR_MINIMIZE | GDK_DECOR_MAXIMIZE);
return TRUE;
case GDK_WINDOW_TYPE_HINT_TOOLBAR:
gdk_window_set_skip_taskbar_hint (window, TRUE);
*decoration = GDK_DECOR_ALL | GDK_DECOR_MINIMIZE | GDK_DECOR_MAXIMIZE;
return TRUE;
case GDK_WINDOW_TYPE_HINT_UTILITY:
return FALSE;
case GDK_WINDOW_TYPE_HINT_SPLASHSCREEN:
*decoration = (GDK_DECOR_ALL | GDK_DECOR_RESIZEH | GDK_DECOR_MENU
| GDK_DECOR_MINIMIZE | GDK_DECOR_MAXIMIZE);
return TRUE;
case GDK_WINDOW_TYPE_HINT_DOCK:
return FALSE;
case GDK_WINDOW_TYPE_HINT_DESKTOP:
return FALSE;
default:
/* Fall thru */
case GDK_WINDOW_TYPE_HINT_NORMAL:
*decoration = GDK_DECOR_ALL;
return TRUE;
}
}
return FALSE;
}
void
gdk_window_set_geometry_hints (GdkWindow *window,
GdkGeometry *geometry,
GdkWindowHints geom_mask)
{
GdkWindowImplWin32 *impl;
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_geometry_hints: %p\n",
GDK_WINDOW_HWND (window)));
impl = GDK_WINDOW_IMPL_WIN32 (GDK_WINDOW_OBJECT (window)->impl);
impl->hint_flags = geom_mask;
impl->hints = *geometry;
if (geom_mask & GDK_HINT_POS)
; /* even the X11 mplementation doesn't care */
if (geom_mask & GDK_HINT_MIN_SIZE)
{
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("... MIN_SIZE: %dx%d\n",
geometry->min_width, geometry->min_height));
}
if (geom_mask & GDK_HINT_MAX_SIZE)
{
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("... MAX_SIZE: %dx%d\n",
geometry->max_width, geometry->max_height));
}
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
if (geom_mask & GDK_HINT_BASE_SIZE)
{
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("... BASE_SIZE: %dx%d\n",
geometry->base_width, geometry->base_height));
}
if (geom_mask & GDK_HINT_RESIZE_INC)
{
GDK_NOTE (MISC, g_print ("... RESIZE_INC: (%d,%d)\n",
geometry->width_inc, geometry->height_inc));
}
if (geom_mask & GDK_HINT_ASPECT)
{
GDK_NOTE (MISC, g_print ("... ASPECT: %g--%g\n",
geometry->min_aspect, geometry->max_aspect));
}
if (geom_mask & GDK_HINT_WIN_GRAVITY)
{
GDK_NOTE (MISC, g_print ("... GRAVITY: %d\n", geometry->win_gravity));
}
update_style_bits (window);
}
void
gdk_window_set_title (GdkWindow *window,
const gchar *title)
{
wchar_t *wtitle;
g_return_if_fail (GDK_IS_WINDOW (window));
Large changes to the Win32 backend, partially made necessary by the 2000-05-02 Tor Lillqvist <tml@iki.fi> Large changes to the Win32 backend, partially made necessary by the changes to the backend-independent internal structures. Attempts to implement similar backing store stuff as on X11. The current (CVS) version of the Win32 backend is *not* as stable as it was before the no-flicker branch was merged. A zipfile with that version is available from http://www.gimp.org/win32/. That should be use by "production" code until this CVS version is usable. (But note, the Win32 backend has never been claimed to be "production quality".) * README.win32: Add the above comment about versions. * gdk/gdkwindow.c: Don't use backing store for now on Win32. * gdk/gdk.def: Update. * gdk/gdkfont.h: Declare temporary Win32-only functions. Will presumably be replaced by some more better mechanism as 1.4 gets closer to release shape. * gdk/makefile.{cygwin,msc}: Update. * gdk/win32/*.c: Correct inclusions of the backend-specific and internal headers. Change code according to changes in these. Use gdk_drawable_*, not gdk_window_* where necessary. * gdk/win32/gdkdnd-win32.c: Use MISC selector for GDK_NOTE, not our old DND. * gdk/win32/gdkdrawable-win32.c (gdk_win32_draw_text): Don't try to interpret single characters as UTF-8. Thanks to Hans Breuer. Use correct function name in warning messages. * gdk/win32/gdkevents-win32.c: Use correct parameter lists for the GSourceFuncs gdk_event_prepare and gdk_event_check. (gdk_event_get_graphics_expose): Do implement, use PeekMessage. Thanks to Hans Breuer. (event_mask_string): Debugging function to print an GdkEventMask. (gdk_pointer_grab): Use it. * gdk/win32/gdkfont-win32.c: The Unicode subrange that the (old) book I used claimed was Hangul actually is CJK Unified Ideographs Extension A. Also, Hangul Syllables were missing. Improve logging. * gdk/win32/gdkgc-win32.c: Largish changes. * gdk/win32/gdkim-win32.c (gdk_set_locale): Use g_win32_getlocale() from GLib, and not setlocale() to get current locale name. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwin32.h: Move stuff from gdkprivate-win32.h to gdkwin32.h, similarily as in the X11 backend. * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Bugfix, assignment was used instead of equals in if test. Thanks to Hans Breuer. * gdk/win32/makefile.{cygwin,msc} * gtk/makefile.{cygwin,msc}: Updates. Better kludge to get the path to the Win32 headers that works also with the mingw compiler. * gtk/gtkstyle.c: Include <string.h>.
2000-05-01 22:06:49 +00:00
g_return_if_fail (title != NULL);
if (GDK_WINDOW_DESTROYED (window))
return;
/* Empty window titles not allowed, so set it to just a period. */
if (!title[0])
title = ".";
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_title: %p: %s\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window), title));
wtitle = g_utf8_to_utf16 (title, -1, NULL, NULL, NULL);
API_CALL (SetWindowTextW, (GDK_WINDOW_HWND (window), wtitle));
g_free (wtitle);
}
void
gdk_window_set_role (GdkWindow *window,
const gchar *role)
{
g_return_if_fail (GDK_IS_WINDOW (window));
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_role: %p: %s\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
(role ? role : "NULL")));
/* XXX */
}
void
gdk_window_set_transient_for (GdkWindow *window,
GdkWindow *parent)
{
HWND window_id, parent_id;
GdkWindowImplWin32 *window_impl = GDK_WINDOW_IMPL_WIN32 (GDK_WINDOW_OBJECT (window)->impl);
g_return_if_fail (GDK_IS_WINDOW (window));
window_id = GDK_WINDOW_HWND (window);
parent_id = parent != NULL ? GDK_WINDOW_HWND (parent) : NULL;
if (GDK_WINDOW_DESTROYED (window) || (parent && GDK_WINDOW_DESTROYED (parent)))
{
if (GDK_WINDOW_DESTROYED (window))
GDK_NOTE (MISC, g_print ("... destroyed!\n"));
else
GDK_NOTE (MISC, g_print ("... owner destroyed!\n"));
return;
}
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
if (((GdkWindowObject *) window)->window_type == GDK_WINDOW_CHILD)
{
GDK_NOTE (MISC, g_print ("... a child window!\n"));
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
return;
}
window_impl->transient_owner = parent;
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
/* This changes the *owner* of the window, despite the misleading
* name. (Owner and parent are unrelated concepts.) At least that's
* what people who seem to know what they talk about say on
* USENET. Search on Google.
*/
SetLastError (0);
if (SetWindowLong (window_id, GWL_HWNDPARENT, (long) parent_id) == 0 &&
GetLastError () != 0)
WIN32_API_FAILED ("SetWindowLong");
}
void
Fix #105497; constify uses of GdkColor. 2004-01-26 Federico Mena Quintero <federico@ximian.com> Fix #105497; constify uses of GdkColor. * gdk/gdkgc.c (gdk_gc_set_rgb_fg_color): Constify. (gdk_gc_set_rgb_bg_color): Constify. (gdk_gc_set_foreground): Constify. * gdk/x11/gdkcursor-x11.c (gdk_cursor_new_from_pixmap): Constify. * gdk/win32/gdkcursor-win32.c (gdk_cursor_new_from_pixmap): Constify. * gdk/linux-fb/gdkcursor-fb.c (gdk_cursor_new_from_pixmap): Constify. * gdk/x11/gdkpixmap-x11.c (gdk_pixmap_create_from_data): Constify. * gdk/win32/gdkpixmap-win32.c (gdk_pixmap_create_from_data): Constify. * gdk/linux-fb/gdkpixmap-fb.c (gdk_pixmap_create_from_data): Constify. * gdk/x11/gdkwindow-x11.c (gdk_window_set_background): Constify. * gdk/win32/gdkwindow-win32.c (gdk_window_set_background): Constify. * gdk/linux-fb/gdkwindow-fb.c (gdk_window_set_background): Constify. * gdk/gdkpango.c (gdk_draw_layout_line_with_colors): Constify. (gdk_draw_layout_with_colors): Constify. * gdk/gdkpixmap.c (gdk_pixmap_colormap_new_from_pixbuf): Constify. (gdk_pixmap_colormap_create_from_xpm): Constify. (gdk_pixmap_create_from_xpm): Constify. (gdk_pixmap_colormap_create_from_xpm_d): Constify. (gdk_pixmap_create_from_xpm_d): Constify. * gtk/gtkcellview.c (gtk_cell_view_set_background_color): Constify. * gtk/gtkclist.c (gtk_clist_set_foreground): Constify. (gtk_clist_set_background): Constify. * gtk/gtkcolorbutton.c (gtk_color_button_new_with_color): Constify. (gtk_color_button_set_color): Constify. * gtk/gtkcolorsel.c (gtk_color_selection_set_current_color): Constify and add a check for color != NULL. (gtk_color_selection_get_current_color): Add a check for color != NULL. (gtk_color_selection_set_previous_color): Constify and add a check for color != NULL. (gtk_color_selection_get_previous_color): Add a check for color != NULL. * gtk/gtkctree.c (gtk_ctree_node_set_foreground): Constify. (gtk_ctree_node_set_background): Constify. * gtk/gtktext.c (gtk_text_insert): Constify. (insert_text_property): Constify. (text_properties_equal): Constify. (new_text_property): Constify. * gtk/gtkwidget.c (gtk_widget_modify_color_component): Constify. (gtk_widget_modify_fg): Constify. (gtk_widget_modify_bg): Constify. (gtk_widget_modify_text): Constify. (gtk_widget_modify_base): Constify.
2004-01-26 20:21:09 +00:00
gdk_window_set_background (GdkWindow *window,
const GdkColor *color)
{
GdkWindowObject *private = (GdkWindowObject *)window;
g_return_if_fail (GDK_IS_WINDOW (window));
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_background: %p: %s\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
_gdk_win32_color_to_string (color)));
private->bg_color = *color;
if (private->bg_pixmap &&
private->bg_pixmap != GDK_PARENT_RELATIVE_BG &&
private->bg_pixmap != GDK_NO_BG)
{
g_object_unref (private->bg_pixmap);
private->bg_pixmap = NULL;
}
}
void
gdk_window_set_back_pixmap (GdkWindow *window,
GdkPixmap *pixmap,
gint parent_relative)
{
GdkWindowObject *private = (GdkWindowObject *)window;
g_return_if_fail (GDK_IS_WINDOW (window));
g_return_if_fail (pixmap == NULL || !parent_relative);
g_return_if_fail (pixmap == NULL || gdk_drawable_get_depth (window) == gdk_drawable_get_depth (pixmap));
if (pixmap && !gdk_drawable_get_colormap (pixmap))
{
g_warning ("gdk_window_set_back_pixmap(): pixmap must have a colormap");
return;
}
if (private->bg_pixmap &&
private->bg_pixmap != GDK_PARENT_RELATIVE_BG &&
private->bg_pixmap != GDK_NO_BG)
g_object_unref (private->bg_pixmap);
if (parent_relative)
{
private->bg_pixmap = GDK_PARENT_RELATIVE_BG;
GDK_NOTE (MISC, g_print (G_STRLOC ": setting background pixmap to parent_relative\n"));
}
else
{
if (pixmap)
{
g_object_ref (pixmap);
private->bg_pixmap = pixmap;
}
else
{
private->bg_pixmap = GDK_NO_BG;
}
}
}
void
gdk_window_set_cursor (GdkWindow *window,
GdkCursor *cursor)
{
GdkWindowImplWin32 *impl;
GdkCursorPrivate *cursor_private;
GdkWindowObject *parent_window;
HCURSOR hcursor;
HCURSOR hprevcursor;
g_return_if_fail (GDK_IS_WINDOW (window));
impl = GDK_WINDOW_IMPL_WIN32 (GDK_WINDOW_OBJECT (window)->impl);
cursor_private = (GdkCursorPrivate*) cursor;
if (GDK_WINDOW_DESTROYED (window))
return;
if (!cursor)
hcursor = NULL;
else
hcursor = cursor_private->hcursor;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_cursor: %p: %p\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
hcursor));
/* First get the old cursor, if any (we wait to free the old one
* since it may be the current cursor set in the Win32 API right
* now).
*/
hprevcursor = impl->hcursor;
if (hcursor == NULL)
impl->hcursor = NULL;
else
{
/* We must copy the cursor as it is OK to destroy the GdkCursor
* while still in use for some window. See for instance
* gimp_change_win_cursor() which calls gdk_window_set_cursor
* (win, cursor), and immediately afterwards gdk_cursor_destroy
* (cursor).
*/
if ((impl->hcursor = CopyCursor (hcursor)) == NULL)
WIN32_API_FAILED ("CopyCursor");
GDK_NOTE (MISC, g_print ("... CopyCursor (%p) = %p\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
hcursor, impl->hcursor));
}
if (impl->hcursor != NULL)
{
/* If the pointer is over our window, set new cursor */
GdkWindow *curr_window = gdk_window_get_pointer (window, NULL, NULL, NULL);
if (curr_window == window)
SetCursor (impl->hcursor);
else
{
/* Climb up the tree and find whether our window is the
* first ancestor that has cursor defined, and if so, set
* new cursor.
*/
GdkWindowObject *curr_window_obj = GDK_WINDOW_OBJECT (curr_window);
while (curr_window_obj &&
!GDK_WINDOW_IMPL_WIN32 (curr_window_obj->impl)->hcursor)
{
curr_window_obj = curr_window_obj->parent;
if (curr_window_obj == GDK_WINDOW_OBJECT (window))
{
SetCursor (impl->hcursor);
break;
}
}
}
}
/* Destroy the previous cursor: Need to make sure it's no longer in
* use before we destroy it, in case we're not over our window but
* the cursor is still set to our old one.
*/
if (hprevcursor != NULL)
{
if (GetCursor () == hprevcursor)
{
/* Look for a suitable cursor to use instead */
hcursor = NULL;
parent_window = GDK_WINDOW_OBJECT (window)->parent;
while (hcursor == NULL)
{
if (parent_window)
{
impl = GDK_WINDOW_IMPL_WIN32 (parent_window->impl);
hcursor = impl->hcursor;
parent_window = parent_window->parent;
}
else
{
hcursor = LoadCursor (NULL, IDC_ARROW);
}
}
SetCursor (hcursor);
}
GDK_NOTE (MISC, g_print ("... DestroyCursor (%p)\n", hprevcursor));
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
API_CALL (DestroyCursor, (hprevcursor));
}
}
void
gdk_window_get_geometry (GdkWindow *window,
gint *x,
gint *y,
gint *width,
gint *height,
gint *depth)
{
g_return_if_fail (window == NULL || GDK_IS_WINDOW (window));
if (!window)
window = _gdk_root;
if (!GDK_WINDOW_DESTROYED (window))
{
RECT rect;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
API_CALL (GetClientRect, (GDK_WINDOW_HWND (window), &rect));
if (window != _gdk_root)
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
{
POINT pt;
GdkWindow *parent = gdk_window_get_parent (window);
pt.x = rect.left;
pt.y = rect.top;
ClientToScreen (GDK_WINDOW_HWND (window), &pt);
ScreenToClient (GDK_WINDOW_HWND (parent), &pt);
rect.left = pt.x;
rect.top = pt.y;
pt.x = rect.right;
pt.y = rect.bottom;
ClientToScreen (GDK_WINDOW_HWND (window), &pt);
ScreenToClient (GDK_WINDOW_HWND (parent), &pt);
rect.right = pt.x;
rect.bottom = pt.y;
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
if (parent == _gdk_root)
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
{
rect.left += _gdk_offset_x;
rect.top += _gdk_offset_y;
rect.right += _gdk_offset_x;
rect.bottom += _gdk_offset_y;
}
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
}
if (x)
*x = rect.left;
if (y)
*y = rect.top;
if (width)
*width = rect.right - rect.left;
if (height)
*height = rect.bottom - rect.top;
if (depth)
*depth = gdk_drawable_get_visual (window)->depth;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_get_geometry: %p: %ldx%ldx%d@%+ld%+ld\n",
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_WINDOW_HWND (window),
rect.right - rect.left, rect.bottom - rect.top,
gdk_drawable_get_visual (window)->depth,
rect.left, rect.top));
}
}
gint
gdk_window_get_origin (GdkWindow *window,
gint *x,
gint *y)
{
gint return_val;
gint tx = 0;
gint ty = 0;
g_return_val_if_fail (GDK_IS_WINDOW (window), 0);
if (!GDK_WINDOW_DESTROYED (window))
{
POINT pt;
pt.x = 0;
pt.y = 0;
ClientToScreen (GDK_WINDOW_HWND (window), &pt);
tx = pt.x;
ty = pt.y;
return_val = 1;
}
else
return_val = 0;
if (x)
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
*x = tx + _gdk_offset_x;
if (y)
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
*y = ty + _gdk_offset_y;
GDK_NOTE (MISC, g_print ("gdk_window_get_origin: %p: %+d%+d\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
tx, ty));
return return_val;
}
gboolean
gdk_window_get_deskrelative_origin (GdkWindow *window,
gint *x,
gint *y)
{
return gdk_window_get_origin (window, x, y);
}
void
gdk_window_get_root_origin (GdkWindow *window,
gint *x,
gint *y)
{
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GdkRectangle rect;
g_return_if_fail (GDK_IS_WINDOW (window));
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
gdk_window_get_frame_extents (window, &rect);
if (x)
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
*x = rect.x;
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
if (y)
*y = rect.y;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_get_root_origin: %p: %+d%+d\n",
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_WINDOW_HWND (window), rect.x, rect.y));
}
2001-03-31 19:33:49 +00:00
void
gdk_window_get_frame_extents (GdkWindow *window,
GdkRectangle *rect)
{
GdkWindowObject *private;
HWND hwnd;
RECT r;
g_return_if_fail (GDK_IS_WINDOW (window));
g_return_if_fail (rect != NULL);
private = GDK_WINDOW_OBJECT (window);
2001-03-31 19:33:49 +00:00
rect->x = 0;
rect->y = 0;
rect->width = 1;
rect->height = 1;
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
/* FIXME: window is documented to be a toplevel GdkWindow, so is it really
* necessary to walk its parent chain?
*/
while (private->parent && ((GdkWindowObject*) private->parent)->parent)
private = (GdkWindowObject*) private->parent;
2001-03-31 19:33:49 +00:00
hwnd = GDK_WINDOW_HWND (window);
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
API_CALL (GetWindowRect, (hwnd, &r));
2001-03-31 19:33:49 +00:00
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
rect->x = r.left + _gdk_offset_x;
rect->y = r.top + _gdk_offset_y;
2001-03-31 19:33:49 +00:00
rect->width = r.right - r.left;
rect->height = r.bottom - r.top;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_get_frame_extents: %p: %ldx%ld@%+ld%+ld\n",
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_WINDOW_HWND (window),
r.right - r.left, r.bottom - r.top,
r.left, r.top));
2001-03-31 19:33:49 +00:00
}
GdkWindow*
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
_gdk_windowing_window_get_pointer (GdkDisplay *display,
GdkWindow *window,
gint *x,
gint *y,
GdkModifierType *mask)
{
GdkWindow *return_val;
POINT screen_point, point;
HWND hwnd, hwndc;
BYTE kbd[256];
g_return_val_if_fail (window == NULL || GDK_IS_WINDOW (window), NULL);
return_val = NULL;
GetCursorPos (&screen_point);
point = screen_point;
ScreenToClient (GDK_WINDOW_HWND (window), &point);
*x = point.x;
*y = point.y;
if (window == _gdk_root)
{
*x += _gdk_offset_x;
*y += _gdk_offset_y;
}
hwnd = WindowFromPoint (screen_point);
if (hwnd != NULL)
{
gboolean done = FALSE;
while (!done)
{
point = screen_point;
ScreenToClient (hwnd, &point);
hwndc = ChildWindowFromPoint (hwnd, point);
if (hwndc == NULL)
done = TRUE;
else if (hwndc == hwnd)
done = TRUE;
else
hwnd = hwndc;
}
return_val = gdk_window_lookup ((GdkNativeWindow) hwnd);
}
else
return_val = NULL;
GetKeyboardState (kbd);
*mask = 0;
if (kbd[VK_SHIFT] & 0x80)
*mask |= GDK_SHIFT_MASK;
if (kbd[VK_CAPITAL] & 0x80)
*mask |= GDK_LOCK_MASK;
if (kbd[VK_CONTROL] & 0x80)
*mask |= GDK_CONTROL_MASK;
if (kbd[VK_MENU] & 0x80)
*mask |= GDK_MOD1_MASK;
if (kbd[VK_LBUTTON] & 0x80)
*mask |= GDK_BUTTON1_MASK;
if (kbd[VK_MBUTTON] & 0x80)
*mask |= GDK_BUTTON2_MASK;
if (kbd[VK_RBUTTON] & 0x80)
*mask |= GDK_BUTTON3_MASK;
return return_val;
}
void
_gdk_windowing_get_pointer (GdkDisplay *display,
GdkScreen **screen,
gint *x,
gint *y,
GdkModifierType *mask)
{
g_return_if_fail (display == _gdk_display);
*screen = _gdk_screen;
_gdk_windowing_window_get_pointer (_gdk_display, _gdk_root, x, y, mask);
}
void
gdk_display_warp_pointer (GdkDisplay *display,
GdkScreen *screen,
gint x,
gint y)
{
g_return_if_fail (display == _gdk_display);
g_return_if_fail (screen == _gdk_screen);
SetCursorPos (x - _gdk_offset_x, y - _gdk_offset_y);
}
GdkWindow*
_gdk_windowing_window_at_pointer (GdkDisplay *display,
gint *win_x,
gint *win_y)
{
GdkWindow *window;
POINT point, pointc;
HWND hwnd, hwndc;
RECT rect;
GetCursorPos (&pointc);
point = pointc;
hwnd = WindowFromPoint (point);
if (hwnd == NULL)
{
window = _gdk_root;
*win_x = pointc.x + _gdk_offset_x;
*win_y = pointc.y + _gdk_offset_y;
return window;
}
ScreenToClient (hwnd, &point);
do {
hwndc = ChildWindowFromPoint (hwnd, point);
ClientToScreen (hwnd, &point);
ScreenToClient (hwndc, &point);
} while (hwndc != hwnd && (hwnd = hwndc, 1));
window = gdk_win32_handle_table_lookup ((GdkNativeWindow) hwnd);
if (window && (win_x || win_y))
{
GetClientRect (hwnd, &rect);
*win_x = point.x - rect.left;
*win_y = point.y - rect.top;
}
GDK_NOTE (MISC, g_print ("_gdk_windowing_window_at_pointer: %+d%+d %p%s\n",
*win_x, *win_y,
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
hwnd,
(window == NULL ? " NULL" : "")));
return window;
}
GdkEventMask
gdk_window_get_events (GdkWindow *window)
{
g_return_val_if_fail (GDK_IS_WINDOW (window), 0);
if (GDK_WINDOW_DESTROYED (window))
return 0;
return GDK_WINDOW_OBJECT (window)->event_mask;
}
void
gdk_window_set_events (GdkWindow *window,
GdkEventMask event_mask)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
/* gdk_window_new() always sets the GDK_STRUCTURE_MASK, so better
* set it here, too. Not that I know or remember why it is
* necessary, will have to test some day.
*/
GDK_WINDOW_OBJECT (window)->event_mask = GDK_STRUCTURE_MASK | event_mask;
}
static void
do_shape_combine_region (GdkWindow *window,
HRGN hrgn,
gint x, gint y)
{
RECT rect;
GetClientRect (GDK_WINDOW_HWND (window), &rect);
_gdk_win32_adjust_client_rect (window, &rect);
OffsetRgn (hrgn, -rect.left, -rect.top);
OffsetRgn (hrgn, x, y);
/* If this is a top-level window, add the title bar to the region */
if (GDK_WINDOW_TYPE (window) == GDK_WINDOW_TOPLEVEL)
{
HRGN tmp = CreateRectRgn (0, 0, rect.right - rect.left, -rect.top);
CombineRgn (hrgn, hrgn, tmp, RGN_OR);
DeleteObject (tmp);
}
SetWindowRgn (GDK_WINDOW_HWND (window), hrgn, TRUE);
}
void
gdk_window_shape_combine_mask (GdkWindow *window,
GdkBitmap *mask,
gint x, gint y)
{
GdkWindowObject *private = (GdkWindowObject *)window;
g_return_if_fail (GDK_IS_WINDOW (window));
if (!mask)
{
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_shape_combine_mask: %p: none\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window)));
SetWindowRgn (GDK_WINDOW_HWND (window), NULL, TRUE);
private->shaped = FALSE;
}
else
{
HRGN hrgn;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_shape_combine_mask: %p: %p\n",
Massive changes. Too many to list here, but I'll try a summary: 2002-02-17 Tor Lillqvist <tml@iki.fi> * gdk/win32/*.c: Massive changes. Too many to list here, but I'll try a summary: 1) Unify GdkPixmap and GdkImage implementation: For each GdkPixmap, allocate a GdkImage, and vice versa. GdkPixmapImplWin32Data has a pointer to the GdkImage. GdkImage::windowing_data is a pointer to the GdkPixmap. This simplifies many pixmap and image related functions a lot, and reduces duplicated code snippets. For instance, there is only one place in gdk/win32 where CreateDIBSection() is called, in the function _gdk_win32_new_pixmap(). Converting a bitmap (GdkPixmap) to a Windows region is almost trivial, with the bitmap bits being readily accessible in the associated GdkImage. All blitting between GdkPixmaps, GdkWindows and GdkImages goes through handled the _gdk_win32_blit() function, which calls different functions to handle the cases of blitting from pixmaps, inside windows (scrolling), or from windows, which all require somewhat different handling. 2) Support 256-color mode. This has long been very broken, now it works more or less OK. Keep the logical palette for each colormap as small as possible while allocating and freeing colors. Select and realize the logical palette associated with a GdkColormap into a DC before drawing or blitting. When the display is in 256-color mode, make it possible for the user to override the size of the palette(s) used with either the GDK_WIN32_MAX_COLORS environment variable, or a -max-colors command line option. It is possible to reduce the palette size all the way down to using just the 16 static colors (which causes the system visual to be of type GDK_VISUAL_STATIC_COLOR. This could possibly be useful if one desperately wants to avoid color flashing. (Note that in order for this to work properly, an as of yet not commited fix to gdkrgb.c is needed.) Handle the palette messages. On WM_PALETTECHANGED, call UpdateColors() for the given window hierarchy. Do this only if a window in some other top-level window hierarchy caused the palette change (realized a palette). Do this max five times in a row (an arbitrarily chosen limit), though, otherwise redraw by generating expose events. On WM_QUERYNEWPALETTE, cause a redraw of the whole window hierarchy by generating GDK_EXPOSE events. 3) Code cleanup in general. For instance, remove the "emulated" X11 structs ColormapStruct, Visual and XStandardColormap. Use the new GDK_DEBUG_* flags for debugging output in the relevant source files. Remove the unused colormap hash table in gdkcolor-win32.c 4) Plug some resource leaks. 2002-02-14 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkdnd-win32.c (gdk_dropfiles_filter): Use g_filename_to_uri() to actually create legal URIs in the text/uri-list data.
2002-02-17 00:25:05 +00:00
GDK_WINDOW_HWND (window),
GDK_WINDOW_HWND (mask)));
/* Convert mask bitmap to region */
hrgn = _gdk_win32_bitmap_to_hrgn (mask);
do_shape_combine_region (window, hrgn, x, y);
private->shaped = TRUE;
}
}
void
gdk_window_input_shape_combine_mask (GdkWindow *window,
GdkBitmap *mask,
gint x,
gint y)
{
g_return_if_fail (GDK_IS_WINDOW (window));
/* Not yet implemented
*
* I don't think there is anything in the Win32 API to directly
* support this. And anyway, as we don't currently support RGBA
* windows, it doesn't really matter.
*
* When we do support RGBA, input shape functionality could probably
* be implemented by saving the input shape region in the per-window
* private data, and then simply checking before generating an input
* event whether the event's coordinates are inside the region.
*/
}
void
gdk_window_set_override_redirect (GdkWindow *window,
gboolean override_redirect)
{
g_return_if_fail (GDK_IS_WINDOW (window));
g_warning ("gdk_window_set_override_redirect not implemented");
}
void
gdk_window_set_accept_focus (GdkWindow *window,
gboolean accept_focus)
{
GdkWindowObject *private;
g_return_if_fail (GDK_IS_WINDOW (window));
private = (GdkWindowObject *)window;
accept_focus = accept_focus != FALSE;
if (private->accept_focus != accept_focus)
private->accept_focus = accept_focus;
}
void
gdk_window_set_focus_on_map (GdkWindow *window,
gboolean focus_on_map)
{
GdkWindowObject *private;
g_return_if_fail (GDK_IS_WINDOW (window));
private = (GdkWindowObject *)window;
focus_on_map = focus_on_map != FALSE;
if (private->focus_on_map != focus_on_map)
private->focus_on_map = focus_on_map;
}
void
gdk_window_set_icon_list (GdkWindow *window,
GList *pixbufs)
{
GdkPixbuf *pixbuf, *big_pixbuf, *small_pixbuf;
gint big_diff, small_diff;
gint big_w, big_h, small_w, small_h;
gint w, h;
gint dw, dh, diff;
HICON small_hicon, big_hicon;
GdkWindowImplWin32 *impl;
gint i, big_i, small_i;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
impl = GDK_WINDOW_IMPL_WIN32 (GDK_WINDOW_OBJECT (window)->impl);
/* ideal sizes for small and large icons */
big_w = GetSystemMetrics (SM_CXICON);
big_h = GetSystemMetrics (SM_CYICON);
small_w = GetSystemMetrics (SM_CXSMICON);
small_h = GetSystemMetrics (SM_CYSMICON);
/* find closest sized icons in the list */
big_pixbuf = NULL;
small_pixbuf = NULL;
big_diff = 0;
small_diff = 0;
i = 0;
while (pixbufs)
{
pixbuf = (GdkPixbuf*) pixbufs->data;
w = gdk_pixbuf_get_width (pixbuf);
h = gdk_pixbuf_get_height (pixbuf);
dw = ABS (w - big_w);
dh = ABS (h - big_h);
diff = dw*dw + dh*dh;
if (big_pixbuf == NULL || diff < big_diff)
{
big_pixbuf = pixbuf;
big_diff = diff;
big_i = i;
}
dw = ABS (w - small_w);
dh = ABS (h - small_h);
diff = dw*dw + dh*dh;
if (small_pixbuf == NULL || diff < small_diff)
{
small_pixbuf = pixbuf;
small_diff = diff;
small_i = i;
}
pixbufs = g_list_next (pixbufs);
i++;
}
/* Create the icons */
big_hicon = _gdk_win32_pixbuf_to_hicon (big_pixbuf);
small_hicon = _gdk_win32_pixbuf_to_hicon (small_pixbuf);
/* Set the icons */
SendMessageW (GDK_WINDOW_HWND (window), WM_SETICON, ICON_BIG,
(LPARAM)big_hicon);
SendMessageW (GDK_WINDOW_HWND (window), WM_SETICON, ICON_SMALL,
(LPARAM)small_hicon);
/* Store the icons, destroying any previous icons */
if (impl->hicon_big)
GDI_CALL (DestroyIcon, (impl->hicon_big));
impl->hicon_big = big_hicon;
if (impl->hicon_small)
GDI_CALL (DestroyIcon, (impl->hicon_small));
impl->hicon_small = small_hicon;
}
void
gdk_window_set_icon (GdkWindow *window,
GdkWindow *icon_window,
GdkPixmap *pixmap,
GdkBitmap *mask)
{
g_return_if_fail (GDK_IS_WINDOW (window));
/* do nothing, use gdk_window_set_icon_list instead */
}
void
gdk_window_set_icon_name (GdkWindow *window,
const gchar *name)
{
/* In case I manage to confuse this again (or somebody else does):
* Please note that "icon name" here really *does* mean the name or
* title of an window minimized as an icon on the desktop, or in the
* taskbar. It has nothing to do with the freedesktop.org icon
* naming stuff.
*/
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
#if 0
/* This is not the correct thing to do. We should keep both the
* "normal" window title, and the icon name. When the window is
* minimized, call SetWindowText() with the icon name, and when the
* window is restored, with the normal window title. Also, the name
* is in UTF-8, so we should do the normal conversion to either wide
* chars or system codepage, and use either the W or A version of
* SetWindowText(), depending on Windows version.
*/
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
API_CALL (SetWindowText, (GDK_WINDOW_HWND (window), name));
#endif
}
GdkWindow *
gdk_window_get_group (GdkWindow *window)
{
g_return_val_if_fail (GDK_IS_WINDOW (window), NULL);
g_return_val_if_fail (GDK_WINDOW_TYPE (window) != GDK_WINDOW_CHILD, NULL);
if (GDK_WINDOW_DESTROYED (window))
return NULL;
g_warning ("gdk_window_get_group not yet implemented");
return NULL;
}
void
gdk_window_set_group (GdkWindow *window,
GdkWindow *leader)
{
g_return_if_fail (GDK_IS_WINDOW (window));
g_return_if_fail (GDK_WINDOW_TYPE (window) != GDK_WINDOW_CHILD);
g_return_if_fail (leader == NULL || GDK_IS_WINDOW (leader));
if (GDK_WINDOW_DESTROYED (window) || GDK_WINDOW_DESTROYED (leader))
return;
g_warning ("gdk_window_set_group not implemented");
}
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
static void
update_single_bit (LONG *style,
BOOL all,
int gdk_bit,
int style_bit)
{
/* all controls the interpretation of gdk_bit -- if all is true, gdk_bit
indicates whether style_bit is off; if all is false, gdk bit indicate whether
style_bit is on */
if ((!all && gdk_bit) || (all && !gdk_bit))
*style |= style_bit;
else
*style &= ~style_bit;
}
static void
update_style_bits (GdkWindow *window)
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
{
GdkWMDecoration decorations;
GdkWMFunction functions;
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
LONG style, exstyle;
BOOL all;
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
RECT rect, before, after;
style = GetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE);
exstyle = GetWindowLong (GDK_WINDOW_HWND (window), GWL_EXSTYLE);
GetClientRect (GDK_WINDOW_HWND (window), &before);
after = before;
AdjustWindowRectEx (&before, style, FALSE, exstyle);
if (get_effective_window_decorations (window, &decorations))
{
all = (decorations & GDK_DECOR_ALL);
update_single_bit (&style, all, decorations & GDK_DECOR_BORDER, WS_BORDER);
update_single_bit (&style, all, decorations & GDK_DECOR_RESIZEH, WS_THICKFRAME);
update_single_bit (&style, all, decorations & GDK_DECOR_TITLE, WS_CAPTION);
update_single_bit (&style, all, decorations & GDK_DECOR_MENU, WS_SYSMENU);
update_single_bit (&style, all, decorations & GDK_DECOR_MINIMIZE, WS_MINIMIZEBOX);
update_single_bit (&style, all, decorations & GDK_DECOR_MAXIMIZE, WS_MAXIMIZEBOX);
}
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
/* XXX this is actually incorrect. The menu entries should be added or removed
from the system menu without affecting the window style. */
if (_gdk_window_get_functions (window, &functions))
{
all = (functions & GDK_DECOR_ALL);
update_single_bit (&style, all, functions & GDK_FUNC_RESIZE, WS_THICKFRAME);
update_single_bit (&style, all, functions & GDK_FUNC_MOVE, WS_THICKFRAME | WS_SYSMENU);
update_single_bit (&style, all, functions & GDK_FUNC_MINIMIZE, WS_MINIMIZE);
update_single_bit (&style, all, functions & GDK_FUNC_MOVE, WS_MAXIMIZE);
update_single_bit (&style, all, functions & GDK_FUNC_CLOSE, WS_SYSMENU);
}
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
SetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE, style);
AdjustWindowRectEx (&after, style, FALSE, exstyle);
GetWindowRect (GDK_WINDOW_HWND (window), &rect);
rect.left += after.left - before.left;
rect.top += after.top - before.top;
rect.right += after.right - before.right;
rect.bottom += after.bottom - before.bottom;
SetWindowPos (GDK_WINDOW_HWND (window), NULL,
rect.left, rect.top,
rect.right - rect.left, rect.bottom - rect.top,
SWP_FRAMECHANGED | SWP_NOACTIVATE |
SWP_NOREPOSITION | SWP_NOZORDER);
}
static GQuark
get_decorations_quark ()
{
static GQuark quark = 0;
if (!quark)
quark = g_quark_from_static_string ("gdk-window-decorations");
return quark;
}
void
gdk_window_set_decorations (GdkWindow *window,
GdkWMDecoration decorations)
{
GdkWMDecoration* decorations_copy;
g_return_if_fail (GDK_IS_WINDOW (window));
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_decorations: %p: %s %s%s%s%s%s%s\n",
GDK_WINDOW_HWND (window),
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
(decorations & GDK_DECOR_ALL ? "clearing" : "setting"),
(decorations & GDK_DECOR_BORDER ? "BORDER " : ""),
(decorations & GDK_DECOR_RESIZEH ? "RESIZEH " : ""),
(decorations & GDK_DECOR_TITLE ? "TITLE " : ""),
(decorations & GDK_DECOR_MENU ? "MENU " : ""),
(decorations & GDK_DECOR_MINIMIZE ? "MINIMIZE " : ""),
(decorations & GDK_DECOR_MAXIMIZE ? "MAXIMIZE " : "")));
decorations_copy = g_malloc (sizeof (GdkWMDecoration));
*decorations_copy = decorations;
g_object_set_qdata_full (G_OBJECT (window), get_decorations_quark (), decorations_copy, g_free);
update_style_bits (window);
}
gboolean
gdk_window_get_decorations (GdkWindow *window,
GdkWMDecoration *decorations)
{
GdkWMDecoration* decorations_set;
g_return_val_if_fail (GDK_IS_WINDOW (window), FALSE);
decorations_set = g_object_get_qdata (G_OBJECT (window), get_decorations_quark ());
if (decorations_set)
*decorations = *decorations_set;
return (decorations_set != NULL);
}
static GQuark
get_functions_quark ()
{
static GQuark quark = 0;
if (!quark)
quark = g_quark_from_static_string ("gdk-window-functions");
return quark;
}
void
gdk_window_set_functions (GdkWindow *window,
GdkWMFunction functions)
{
GdkWMFunction* functions_copy;
g_return_if_fail (GDK_IS_WINDOW (window));
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_functions: %p: %s %s%s%s%s%s\n",
GDK_WINDOW_HWND (window),
Remove unnecessary includes. 2005-11-09 Tor Lillqvist <tml@novell.com> * gdk/win32/gdkwindow-win32.c: Remove unnecessary includes. (_gdk_win32_get_adjusted_client_rect): Remove this two-line function which was used only in one place. (get_outer_rect): Use _gdk_win32_adjust_client_rect(). (gdk_window_set_geometry_hints): If we have identical minimum and maximum size hints, remove the resize and maximize decorations/functions. (#104514) If we have a maximum size hint, remove the maximize decoration/function but ensure the resize decoration/function is available. Otherwise ensure both resize and maximize decorations/functions are there. (set_or_clear_style_bits): Factored out common code from gdk_window_set_decorations() and gdk_window_set_functions(). Hack the window style setting once more: Only touch the window style bits corresponding to the GdkWMDecoration or GdkWMFunction parameter bitmasks. Hopefully this finally is the correct thing to do. We used to clear all other bits than those that were being set, or set all other bits than those that were being cleared. Take into account that adding or removing decorations leaves the window's outer size unchanged, i.e., the client area's size and position change. This is apparently not what we want, so change also the window's (outer) position and size appropriately so that the client area's position and size stay constant. gtk-demo's color selector dialog is now non-resizable like on X11 (I tested with metacity in GNOME). Torn off menus are shrinkable vertically but have a maximum size, and are not maximizable or minimizable, like on X11. (gdk_window_set_decorations, gdk_window_set_functions): Let set_or_clear_decorations() do most of the job. * gdk/win32/gdkprivate-win32.h: Remove declaration of _gdk_win32_get_adjusted_client_rect().
2005-11-09 12:35:56 +00:00
(functions & GDK_FUNC_ALL ? "clearing" : "setting"),
(functions & GDK_FUNC_RESIZE ? "RESIZE " : ""),
(functions & GDK_FUNC_MOVE ? "MOVE " : ""),
(functions & GDK_FUNC_MINIMIZE ? "MINIMIZE " : ""),
(functions & GDK_FUNC_MAXIMIZE ? "MAXIMIZE " : ""),
(functions & GDK_FUNC_CLOSE ? "CLOSE " : "")));
functions_copy = g_malloc (sizeof (GdkWMFunction));
*functions_copy = functions;
g_object_set_qdata_full (G_OBJECT (window), get_functions_quark (), functions_copy, g_free);
update_style_bits (window);
}
gboolean
_gdk_window_get_functions (GdkWindow *window,
GdkWMFunction *functions)
{
GdkWMDecoration* functions_set;
functions_set = g_object_get_qdata (G_OBJECT (window), get_functions_quark ());
if (functions_set)
*functions = *functions_set;
return (functions_set != NULL);
}
static void
QueryTree (HWND hwnd,
HWND **children,
gint *nchildren)
{
guint i, n;
HWND child;
n = 0;
do {
if (n == 0)
child = GetWindow (hwnd, GW_CHILD);
else
child = GetWindow (child, GW_HWNDNEXT);
if (child != NULL)
n++;
} while (child != NULL);
if (n > 0)
{
*children = g_new (HWND, n);
for (i = 0; i < n; i++)
{
if (i == 0)
child = GetWindow (hwnd, GW_CHILD);
else
child = GetWindow (child, GW_HWNDNEXT);
*children[i] = child;
}
}
}
static void
gdk_propagate_shapes (HANDLE win,
gboolean merge)
{
RECT emptyRect;
HRGN region, childRegion;
HWND *list = NULL;
gint i, num;
SetRectEmpty (&emptyRect);
region = CreateRectRgnIndirect (&emptyRect);
if (merge)
GetWindowRgn (win, region);
QueryTree (win, &list, &num);
if (list != NULL)
{
WINDOWPLACEMENT placement;
placement.length = sizeof (WINDOWPLACEMENT);
/* go through all child windows and combine regions */
for (i = 0; i < num; i++)
{
GetWindowPlacement (list[i], &placement);
Large changes to the Win32 backend, partially made necessary by the 2000-05-02 Tor Lillqvist <tml@iki.fi> Large changes to the Win32 backend, partially made necessary by the changes to the backend-independent internal structures. Attempts to implement similar backing store stuff as on X11. The current (CVS) version of the Win32 backend is *not* as stable as it was before the no-flicker branch was merged. A zipfile with that version is available from http://www.gimp.org/win32/. That should be use by "production" code until this CVS version is usable. (But note, the Win32 backend has never been claimed to be "production quality".) * README.win32: Add the above comment about versions. * gdk/gdkwindow.c: Don't use backing store for now on Win32. * gdk/gdk.def: Update. * gdk/gdkfont.h: Declare temporary Win32-only functions. Will presumably be replaced by some more better mechanism as 1.4 gets closer to release shape. * gdk/makefile.{cygwin,msc}: Update. * gdk/win32/*.c: Correct inclusions of the backend-specific and internal headers. Change code according to changes in these. Use gdk_drawable_*, not gdk_window_* where necessary. * gdk/win32/gdkdnd-win32.c: Use MISC selector for GDK_NOTE, not our old DND. * gdk/win32/gdkdrawable-win32.c (gdk_win32_draw_text): Don't try to interpret single characters as UTF-8. Thanks to Hans Breuer. Use correct function name in warning messages. * gdk/win32/gdkevents-win32.c: Use correct parameter lists for the GSourceFuncs gdk_event_prepare and gdk_event_check. (gdk_event_get_graphics_expose): Do implement, use PeekMessage. Thanks to Hans Breuer. (event_mask_string): Debugging function to print an GdkEventMask. (gdk_pointer_grab): Use it. * gdk/win32/gdkfont-win32.c: The Unicode subrange that the (old) book I used claimed was Hangul actually is CJK Unified Ideographs Extension A. Also, Hangul Syllables were missing. Improve logging. * gdk/win32/gdkgc-win32.c: Largish changes. * gdk/win32/gdkim-win32.c (gdk_set_locale): Use g_win32_getlocale() from GLib, and not setlocale() to get current locale name. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwin32.h: Move stuff from gdkprivate-win32.h to gdkwin32.h, similarily as in the X11 backend. * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Bugfix, assignment was used instead of equals in if test. Thanks to Hans Breuer. * gdk/win32/makefile.{cygwin,msc} * gtk/makefile.{cygwin,msc}: Updates. Better kludge to get the path to the Win32 headers that works also with the mingw compiler. * gtk/gtkstyle.c: Include <string.h>.
2000-05-01 22:06:49 +00:00
if (placement.showCmd == SW_SHOWNORMAL)
{
childRegion = CreateRectRgnIndirect (&emptyRect);
GetWindowRgn (list[i], childRegion);
CombineRgn (region, region, childRegion, RGN_OR);
DeleteObject (childRegion);
}
}
SetWindowRgn (win, region, TRUE);
gdk/win32/gdkprivate-win32.h Rename all global variables and functions to 2002-11-12 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkprivate-win32.h * gdk/win32/*.c: Rename all global variables and functions to start with underscore. Merge from stable: More work on the Win32 backend. The cause of some scrolling problems was that SetWindowPos() and ScrollWindowEx() don't blit those parts of the window they think are invalid. As we didn't keep Windows's update region in synch with GDK's, Windows thought those areas that in fact had been updated were invalid. Calling ValidateRgn() in _gdk_windowing_window_queue_antiexpose() seems to be an elegant and efficient solution, removing from Windows's update region those areas we are about to repaint proactively. In some cases garbage leftover values were used for the clip origin in GdkGCWin32. This showed up as odd blank areas around the pixmaps included in the Text Widget in gtk-demo. Having the clip region either as a GdkRegion or a HRGN in GdkGCWin32 was unnecessary, it's better to just use a HRGN. The translation and antiexpose queue handling in gdkgeometry-win32.c seems unnecessary (and not implementable in the same way as on X11 anyway, no serial numbers) on Windows, ifdeffed out. Don't (try to) do guffaw scrolling as there is no static window gravity on Windows. Guffaw scrolling would be unnecessary anyway, as there is the ScrollWindow() API. This improves the behaviour of the Text Widget demo in gtk-demo a lot. But I have no idea how the lack of static win gravity should be handled in other places where the X11 code uses it. Especially _gdk_window_move_resize_child(). There is still some problem in expose handling. By moving an obscuring window back and forth over testgtk's main window, for instance, every now and then you typically get narrow vertical or horizontal strips of pixels that haven't been properly redrawn after being exposed. A fencepost error somewhere? Otherwise, all of testgtk and gtk-demo except "big windows" now seem to work pretty well. Bug #79720 should be fixed now. * gdk/win32/gdkcolor-win32.c (gdk_win32_color_to_string, gdk_win32_print_paletteentries, gdk_win32_print_system_palette, gdk_win32_print_hpalette) * gdk/win32/gdkdrawable-win32.c (gdk_win32_drawable_description) * gdk/win32/gdkevents-win32.c (gdk_win32_message_name): Move all debugging helper functions to gdkmain-win32.c. * gdk/win32/gdkdrawable-win32.c (_gdk_win32_draw_tiles): Rewrite. Make static. Must take tile origin parameters, too. (gdk_win32_draw_rectangle): Pass the tile/stipple origin to _gdk_win32_draw_tiles(). Remove #if 0 code. (blit_inside_window): Don't call ScrollDC(), that didn't work at all like I thought. A simple call to BitBlt() is enough. * gdk/win32/gdkevents-win32.c (gdk_event_translate) Remove unused latin_locale_loaded variable. (_gdk_win32_get_next_tick): New function. Used to make sure timestamps of events are always increasing, both in events generated from the window procedure and in events gotten via PeekMessage(). Not sure whether this is actually useful, but it seemed as a good idea. (real_window_procedure): Don't use a local GdkEventPrivate variable. Don't attempt any compression of configure or expose events here, handled elsewhere. (erase_background): Accumulate window offsets when traversing up the parent chain for GDK_PARENT_RELATIVE_BG, in order to get correct alignment of background pixmaps. Don't fill with BLACK_BRUSH if GDK_NO_BG. (gdk_event_get_graphics_expose): A bit more verbose debugging output. (gdk_event_translate): Use _gdk_win32_get_next_tick(). In the WM_PAINT handler, don't check for empty update rect. When we get a WM_PAINT, the update region isn't empty. And if it for some strange reason is, that will be handled later anyway. Call GetUpdateRgn() before calling BeginPaint() and EndPaint() (which empty the update region). * gdk/win32/gdkdnd-win32.c * gdk/win32/gdkinput-win32.c: Use _gdk_win32_get_next_tick(). * gdk/win32/gdkfont-win32.c: Use %p to print HFONTs. (gdk_text_size): Remove, unused. * gdk/win32/gdkgc-win32.c: Set clip origins to zero when appropriate. (gdk_gc_copy): Increase refcount on colormap if present. (gdk_win32_hdc_get): Handle just hcliprgn. If we have a stipple, combine it with clip region after selecting into the DC. (_gdk_win32_bitmap_to_hrgn): Rename from _gdk_win32_bitmap_to_region. (_gdk_win3_gdkregion_to_hrgn): New function, code snippet extracted from gdk_win32_hdc_get(). * gdk/win32/gdkgeometry-win32.c: Ifdef out the translate_queue handling. (gdk_window_copy_area_scroll): Increase clipRect to avoid ScrollWindowEx() not scrolling pixels it thinks are invalid. Scroll also children with the ScrollWindowEx() call. No need to call gdk_window_move() on the children. (gdk_window_scroll): Don't do guffaw scrolling. (gdk_window_compute_position): Fix typo, used win32_y where x was intended. (gdk_window_premove, gdk_window_postmove, gdk_window_clip_changed): Add debugging output. (_gdk_windowing_window_queue_antiexpose): Just call ValidateRgn() on the region. (_gdk_window_process_expose): No use for the serial number parameter now. Instead of a rectangle, take a region parameter, as Windows gives us one in WM_PAINT. * gdk/win32/gdkmain-win32.c (_gdk_win32_lbstyle_to_string, _gdk_win32_pstype_to_string, _gdk_win32_psstyle_to_string, _gdk_win32_psendcap_to_string, _gdk_win32_psjoin_to_string, _gdk_win32_rect_to_string, _gdk_win32_gdkrectangle_to_string, _gdk_win32_gdkregion_to_string): New debugging functions. (static_printf): Helper function for the above. sprintfs into a static circular buffer, return value should be used "soon". * gdk/win32/gdkwindow-win32.c (gdk_propagate_shapes): Plug memory leak, free list after use. (gdk_window_gravity_works): Remove, we know that there is no such thing on Windows. (gdk_window_set_static_bit_gravity, gdk_window_set_static_win_gravity): Ditto, remove, they didn't do anything anyway. (_gdk_windowing_window_init, gdk_window_foreign_new): Call _gdk_window_init_position() like in the X11 backend. (gdk_window_reparent): Don't call the now nonexistent gdk_window_set_static_win_gravity(). No idea what should be done instead. (gdk_window_get_geometry): The returned x and y should be relative to parent. Used to be always zero.. (gdk_window_set_static_gravities): Return FALSE if trying to set static gravity. * gdk/win32/gdkprivate-win32.h: Drop the clip_region field from GdkGCWin32. Only use the HRGN hcliprgn. Declare new functions. * gdk/win32/*.c: Use new debugging functions. * gdk/win32/rc/gdk.rc.in: Update copyright year.
2002-11-12 22:17:48 +00:00
g_free (list);
}
else
DeleteObject (region);
}
void
gdk_window_set_child_shapes (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
gdk_propagate_shapes (GDK_WINDOW_HWND (window), FALSE);
}
void
gdk_window_merge_child_shapes (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
gdk_propagate_shapes (GDK_WINDOW_HWND (window), TRUE);
}
void
gdk_window_set_child_input_shapes (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
/* Not yet implemented. See comment in
* gdk_window_input_shape_combine_mask().
*/
}
void
gdk_window_merge_child_input_shapes (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
/* Not yet implemented. See comment in
* gdk_window_input_shape_combine_mask().
*/
}
gboolean
gdk_window_set_static_gravities (GdkWindow *window,
gboolean use_static)
{
g_return_val_if_fail (GDK_IS_WINDOW (window), FALSE);
return !use_static;
}
void
gdk_window_begin_resize_drag (GdkWindow *window,
GdkWindowEdge edge,
gint button,
gint root_x,
gint root_y,
guint32 timestamp)
{
WPARAM winedge;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
/* Tell Windows to start interactively resizing the window by pretending that
* the left pointer button was clicked in the suitable edge or corner. This
* will only work if the button is down when this function is called, and
* will only work with button 1 (left), since Windows only allows window
* dragging using the left mouse button.
*/
if (button != 1)
return;
/* Must break the automatic grab that occured when the button was
* pressed, otherwise it won't work.
*/
gdk_display_pointer_ungrab (_gdk_display, 0);
switch (edge)
{
case GDK_WINDOW_EDGE_NORTH_WEST:
winedge = HTTOPLEFT;
break;
case GDK_WINDOW_EDGE_NORTH:
winedge = HTTOP;
break;
case GDK_WINDOW_EDGE_NORTH_EAST:
winedge = HTTOPRIGHT;
break;
case GDK_WINDOW_EDGE_WEST:
winedge = HTLEFT;
break;
case GDK_WINDOW_EDGE_EAST:
winedge = HTRIGHT;
break;
case GDK_WINDOW_EDGE_SOUTH_WEST:
winedge = HTBOTTOMLEFT;
break;
case GDK_WINDOW_EDGE_SOUTH:
winedge = HTBOTTOM;
break;
case GDK_WINDOW_EDGE_SOUTH_EAST:
default:
winedge = HTBOTTOMRIGHT;
break;
}
DefWindowProcW (GDK_WINDOW_HWND (window), WM_NCLBUTTONDOWN, winedge,
MAKELPARAM (root_x - _gdk_offset_x, root_y - _gdk_offset_y));
}
void
gdk_window_begin_move_drag (GdkWindow *window,
gint button,
gint root_x,
gint root_y,
guint32 timestamp)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
/* Tell Windows to start interactively moving the window by pretending that
* the left pointer button was clicked in the titlebar. This will only work
* if the button is down when this function is called, and will only work
* with button 1 (left), since Windows only allows window dragging using the
* left mouse button.
*/
if (button != 1)
return;
/* Must break the automatic grab that occured when the button was pressed,
* otherwise it won't work.
*/
gdk_display_pointer_ungrab (_gdk_display, 0);
DefWindowProcW (GDK_WINDOW_HWND (window), WM_NCLBUTTONDOWN, HTCAPTION,
MAKELPARAM (root_x - _gdk_offset_x, root_y - _gdk_offset_y));
}
/*
* Setting window states
*/
void
gdk_window_iconify (GdkWindow *window)
{
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
HWND old_active_window;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_iconify: %p: %s\n",
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (((GdkWindowObject *) window)->state)));
if (GDK_WINDOW_IS_MAPPED (window))
{
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
old_active_window = GetActiveWindow ();
ShowWindow (GDK_WINDOW_HWND (window), SW_MINIMIZE);
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
if (old_active_window != GDK_WINDOW_HWND (window))
SetActiveWindow (old_active_window);
}
else
{
gdk_synthesize_window_state (window,
0,
GDK_WINDOW_STATE_ICONIFIED);
}
}
void
gdk_window_deiconify (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_deiconify: %p: %s\n",
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (((GdkWindowObject *) window)->state)));
if (GDK_WINDOW_IS_MAPPED (window))
{
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
show_window_internal (window, FALSE, TRUE);
}
else
{
gdk_synthesize_window_state (window,
GDK_WINDOW_STATE_ICONIFIED,
0);
}
}
void
gdk_window_stick (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
/* FIXME: Do something? */
}
void
gdk_window_unstick (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
/* FIXME: Do something? */
}
void
gdk_window_maximize (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_maximize: %p: %s\n",
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (((GdkWindowObject *) window)->state)));
if (GDK_WINDOW_IS_MAPPED (window))
ShowWindow (GDK_WINDOW_HWND (window), SW_MAXIMIZE);
else
gdk_synthesize_window_state (window,
0,
GDK_WINDOW_STATE_MAXIMIZED);
}
void
gdk_window_unmaximize (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_unmaximize: %p: %s\n",
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (((GdkWindowObject *) window)->state)));
if (GDK_WINDOW_IS_MAPPED (window))
ShowWindow (GDK_WINDOW_HWND (window), SW_RESTORE);
else
gdk_synthesize_window_state (window,
GDK_WINDOW_STATE_MAXIMIZED,
0);
}
typedef struct _FullscreenInfo FullscreenInfo;
struct _FullscreenInfo
{
RECT r;
guint hint_flags;
LONG style;
};
void
gdk_window_fullscreen (GdkWindow *window)
{
gint width, height;
FullscreenInfo *fi;
GdkWindowObject *private = (GdkWindowObject *) window;
g_return_if_fail (GDK_IS_WINDOW (window));
fi = g_new (FullscreenInfo, 1);
if (!GetWindowRect (GDK_WINDOW_HWND (window), &(fi->r)))
g_free (fi);
else
{
GdkWindowImplWin32 *impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
gdk/win32/gdkinput-win32.h Drop the GdkEvent* parameter, it wasn't used. 2003-08-07 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkinput-win32.h * gdk/win32/gdkinput-win32.c (_gdk_input_configure_event, _gdk_input_enter_event): Drop the GdkEvent* parameter, it wasn't used. * gdk/win32/gdkevents-win32.c (gdk_event_translate): Adapt caller accordingly, in fact an uninitialised variable was dereferenced. [Win32] Add support for multiple monitors. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkglobals-win32.c: New global variables for multiple-monitor info: _gdk_num_monitors, _gdk_monitors, and _gdk_offset_x and _gdk_offset_y. * gdk/win32/gdkdisplay-win32.c (count_monitor, enum_monitor): New functions, enumeration functions passed to EnumDisplayMonitors(). (gdk_display_open): If the EnumDisplayMonitors() and GetMonitorInfo() API is present (on Win98, Win2000 and newer), use if to find out monitor info. Calculate the offset between Win32 coordinates (relative to the primary monitor's origin (and thus negative on monitors to the left of or above it), and GDK's (visible coordinates should be non-negative). * gdk/win32/gdkscreen-win32 (gdk_screen_get_n_monitors, gdk_screen_get_monitor_geometry): Use information collected above. (gdk_window_move, gdk_window_move_resize_window_get_geometry): Subtract _gdk_offset_{x,y} from GDK root window coordinates. (gdk_window_get_geometry, gdk_window_get_origin, gdk_window_get_frame_extents): For top-level windows, add _gdk_offset_{x,y} to GDK root window coordinates Still need to handle multiple monitors in gdk_window_fullscreen(). Probably should make the window fullscreen on the monitor where the cursor is? * gdk/win32/gdkevents-win32.c: Add _gdk_offset_{x,y} to all GDK root window coordinates in GdkEvents. [Win32] Fix geometry hint handling. Add support for resize increment and base size, and aspect ratio geometry hints. The "gridded geometry" test in testgtk now works beautifully. * gdk/win32/gdkwindow-win32.c (gdk_window_set_geometry_hints): Turns out this function shouldn't actually ever modify the window's size, just store the hints. (Old code kept for a while inside #if 0.) (gdk_window_set_hints): Remove presumably broken code that handles the position hints, this function is obsolete anyway. * gdk/win32/gdkevents-win32.c: Drop the current_{x,y}_root variables, not used. (adjust_drag): New function, used to implement resize increment hints. (gdk_event_translate): Handle WM_SIZING, implement resize increment and base size, and aspect ratio geometry hints here. The WM_GETMINMAXINFO handler takes care of the minimum and maximum size hints as before. Fix the WM_GETMINMAXINFO handler to take into account window decorations. No need to modify the ptMaxPosition and ptMaxSize fields in the MINMAXINFO struct, the defaults are fine. * gdk/win32/gdkprivate-win32.h * gdk/win32/gdkwindow-win32.c (_gdk_win32_adjust_client_rect, _gdk_win32_get_adjusted_client_rect): New helper functions.
2003-08-07 22:17:18 +00:00
width = GetSystemMetrics (SM_CXSCREEN);
height = GetSystemMetrics (SM_CYSCREEN);
/* remember for restoring */
fi->hint_flags = impl->hint_flags;
impl->hint_flags &= ~GDK_HINT_MAX_SIZE;
g_object_set_data (G_OBJECT (window), "fullscreen-info", fi);
fi->style = GetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE);
SetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE,
(fi->style & ~WS_OVERLAPPEDWINDOW) | WS_POPUP);
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), HWND_TOP,
0, 0, width, height,
SWP_NOCOPYBITS | SWP_SHOWWINDOW));
gdk_synthesize_window_state (window, 0, GDK_WINDOW_STATE_FULLSCREEN);
}
}
void
gdk_window_unfullscreen (GdkWindow *window)
{
FullscreenInfo *fi;
GdkWindowObject *private = (GdkWindowObject *) window;
g_return_if_fail (GDK_IS_WINDOW (window));
fi = g_object_get_data (G_OBJECT (window), "fullscreen-info");
if (fi)
{
GdkWindowImplWin32 *impl = GDK_WINDOW_IMPL_WIN32 (private->impl);
impl->hint_flags = fi->hint_flags;
SetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE, fi->style);
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window), HWND_NOTOPMOST,
fi->r.left, fi->r.top,
fi->r.right - fi->r.left, fi->r.bottom - fi->r.top,
SWP_NOCOPYBITS | SWP_SHOWWINDOW));
g_object_set_data (G_OBJECT (window), "fullscreen-info", NULL);
g_free (fi);
gdk_synthesize_window_state (window, GDK_WINDOW_STATE_FULLSCREEN, 0);
}
}
void
gdk_window_set_keep_above (GdkWindow *window, gboolean setting)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
if (GDK_WINDOW_IS_MAPPED (window))
{
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window),
setting ? HWND_TOPMOST : HWND_NOTOPMOST,
0, 0, 0, 0,
SWP_NOSIZE | SWP_NOMOVE | SWP_NOACTIVATE));
}
else
gdk_synthesize_window_state (window,
setting ? GDK_WINDOW_STATE_BELOW : GDK_WINDOW_STATE_ABOVE,
setting ? GDK_WINDOW_STATE_ABOVE : 0);
}
void
gdk_window_set_keep_below (GdkWindow *window, gboolean setting)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
if (GDK_WINDOW_IS_MAPPED (window))
{
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window),
setting ? HWND_BOTTOM : HWND_NOTOPMOST,
0, 0, 0, 0,
SWP_NOSIZE | SWP_NOMOVE | SWP_NOACTIVATE));
}
else
gdk_synthesize_window_state (window,
setting ? GDK_WINDOW_STATE_ABOVE : GDK_WINDOW_STATE_BELOW,
setting ? GDK_WINDOW_STATE_BELOW : 0);
}
void
gdk_window_focus (GdkWindow *window,
guint32 timestamp)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_focus: %p: %s\n",
Merge from stable: 2002-12-09 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/win32/gdkdrawable-win32.c (generic_draw): Don't leak stipple_gc. More checks for errors. Use correct ternary ROP when blitting the foreground into the tile pixmap onto those pixels where the stipple is set. (I didn't notice that I had used the wrong one, as it didn't matter on Win2k, where DIB sections apparently are zeroed upon creation. But on Win98 they have random initial contents. Thanks to Hans Breuer for reporting this.) (gdk_win32_draw_rectangle, gdk_win32_draw_arc, gdk_win32_draw_polygon): Don't pass the LINE_ATTRIBUTES bits to generic_draw() if drawing a filled figure. * gdk/win32/gdkmain-win32.c (_gdk_win32_print_dc): Minor cosmetics. (_gdk_win32_gcvalues_mask_to_string): Initialize buffer as empty. (_gdk_win32_window_state_to_string): New debugging output helper function. * gdk/win32/gdkevents-win32.c: Minor debugging output changes. (gdk_event_translate): Ignore the WM_SHOWWINDOW/SW_OTHERUNZOOM or SW_OTHERZOOM messages. Do not generate a GDK_UNMAP event for WM_SIZE/SIZE_MINIMIZED messages, they do not really corrspond to unmapping on X11. Set window state correctly for all three of SIZE_{MINIMIZED,MAXIMIZED,RESTORED}. A maximized and then iconified ("minimized" in Windows terminology) window still has the "maximized" property, i.e. when deiconified, it will reappear as maximized. (#10557) * gdk/win32/gdkprivate-win32.h: Declare new function. (WIN32_API_FAILED, WIN32_GDI_FAILED, OTHER_API_FAILED): Don't use __PRETTY_FUNCTION__ if __GNUC__ >= 3, to avoid warning message. * gdk/win32/gdkwindow-win32.c (show_window_internal): Handle more situations. Add parameter to tell whether deiconifying. Code reorg: Return early when appropriate instead of using nested if statements. If just deiconifying without raising, restore active window. (#10557) (gdk_window_hide, gdk_window_withdraw, gdk_window_iconify, gdk_window_deiconify, gdk_window_maximize, gdk_window_unmaximize, gdk_window_focus): Use _gdk_win32_window_state_to_string() in debugging output. (gdk_window_iconify): Restore active window after calling ShowWindow(). Otherwise the "next" window gets activated. (gdk_window_stick, gdk_window_unstick): Don't output any warnings. (gdk_window_set_transient_for): Rewrite. Just call SetWindowLong() with GWL_HWNDPARENT, which despite its name sets the *owner* window, which should be exactly what we want. The PSDK documentation is said to be misleading. testgtk's modal window test now works much better. (#50586)
2002-12-09 00:43:42 +00:00
GDK_WINDOW_HWND (window),
_gdk_win32_window_state_to_string (((GdkWindowObject *) window)->state)));
Bypass calls to the grab/ungrab functions in gdkinput-win32.c, as they 2002-03-06 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkevents-win32.c (gdk_pointer_grab, gdk_pointer_ungrab): Bypass calls to the grab/ungrab functions in gdkinput-win32.c, as they don't effectively do anything anyway. (build_key_event_state): Set MOD2 bit if key is in AltGr group. (gdk_event_translate): Call _gdk_windowing_window_get_offsets() each time the offsets are used, on the window they refer to, instead of once in the beginning. The window in question might change due to event propagation. (gdk_event_translate): Set key event group to 0 for non-AltGr keys. Do set the SHIFT bit in the key event state also for ISO_Left_Tab. Now backtabbing finally works again. * gdk/win32/gdkunput-win32.c: Minor debugging output change. * gdk/win32/gdkkeys-win32.c (update_keymap): Set VK_TAB mapping to Gdk_Tab and GDK_ISO_Left_Tab, like on X11. (gdk_keymap_translate_keyboard_state): Add similar code as in the non-XKB case on X11 to generate a more correct consumed_modifiers. Add debugging output. (gdk_keyval_name): Use the U+xxxx format for UCS characters encoded as keyvals. Never return NULL, but hex number representation if keyval not in table. * gdk/win32/gdkwindow-win32.c (gdk_window_focus): Call SetFocus(). Doesn't seem to have any harmful effect, and probably is close to what this function is supposed to do. But it didn't fix GtkCombo as I had hoped. (gdk_window_set_type_hint): Don't intern the _NET_WM_* atoms that weren't used and wouldn't have any meaning on Win32 anyway.
2002-03-06 00:36:08 +00:00
if (((GdkWindowObject *) window)->state & GDK_WINDOW_STATE_MAXIMIZED)
ShowWindow (GDK_WINDOW_HWND (window), SW_SHOWMAXIMIZED);
else
ShowWindow (GDK_WINDOW_HWND (window), SW_SHOWNORMAL);
Bypass calls to the grab/ungrab functions in gdkinput-win32.c, as they 2002-03-06 Tor Lillqvist <tml@iki.fi> * gdk/win32/gdkevents-win32.c (gdk_pointer_grab, gdk_pointer_ungrab): Bypass calls to the grab/ungrab functions in gdkinput-win32.c, as they don't effectively do anything anyway. (build_key_event_state): Set MOD2 bit if key is in AltGr group. (gdk_event_translate): Call _gdk_windowing_window_get_offsets() each time the offsets are used, on the window they refer to, instead of once in the beginning. The window in question might change due to event propagation. (gdk_event_translate): Set key event group to 0 for non-AltGr keys. Do set the SHIFT bit in the key event state also for ISO_Left_Tab. Now backtabbing finally works again. * gdk/win32/gdkunput-win32.c: Minor debugging output change. * gdk/win32/gdkkeys-win32.c (update_keymap): Set VK_TAB mapping to Gdk_Tab and GDK_ISO_Left_Tab, like on X11. (gdk_keymap_translate_keyboard_state): Add similar code as in the non-XKB case on X11 to generate a more correct consumed_modifiers. Add debugging output. (gdk_keyval_name): Use the U+xxxx format for UCS characters encoded as keyvals. Never return NULL, but hex number representation if keyval not in table. * gdk/win32/gdkwindow-win32.c (gdk_window_focus): Call SetFocus(). Doesn't seem to have any harmful effect, and probably is close to what this function is supposed to do. But it didn't fix GtkCombo as I had hoped. (gdk_window_set_type_hint): Don't intern the _NET_WM_* atoms that weren't used and wouldn't have any meaning on Win32 anyway.
2002-03-06 00:36:08 +00:00
SetFocus (GDK_WINDOW_HWND (window));
}
void
gdk_window_set_modal_hint (GdkWindow *window,
gboolean modal)
{
GdkWindowObject *private;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
private = (GdkWindowObject*) window;
private->modal_hint = modal;
#if 0
/* Not sure about this one.. -- Cody */
if (GDK_WINDOW_IS_MAPPED (window))
API_CALL (SetWindowPos, (GDK_WINDOW_HWND (window),
modal ? HWND_TOPMOST : HWND_NOTOPMOST,
0, 0, 0, 0,
SWP_NOMOVE | SWP_NOSIZE));
#endif
}
void
gdk_window_set_skip_taskbar_hint (GdkWindow *window,
gboolean skips_taskbar)
{
Merge from stable: 2003-12-14 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/gdkdisplaymanager.c: Mark default_display static. * gdk/win32/gdkdnd-win32.c: Mark current_dest_drag static. * gdk/win32/gdkkeys-win32.c: Disable some overly verbose debugging output. * gdk/win32/gdkevents-win32.c: Clean up the debugging output from --gdk-debug=events. In general, output just one line of debugging output for each Windows message, plus one line for each GDK event generated. Indent all lines according to window procedure nesting level. (inner_window_procedure): Rename from real_window_procedure. (find_real_window_for_grabbed_mouse_event): Don't get misled when the point is in the non-client (decoration) area of the window returned by WindowFromPoint(). Return the root window in that case. (build_pointer_event_state): Test also MK_XBUTTON1 and MK_XBUTTON2 (buttons 4 and 5). (synthesize_enter_event): Track the mouse leaving the window in the event being generated, not the one mentioned in the Windows message. (propagate): Test for NULL parent earlier. Improves event generation from a grabbed pointer. Part of fix for #107320. (handle_stuff_while_moving_or_resizing): New function, to dispatch the main loop (once). (resize_timer_proc): New function, set to be called by an inerval timer during resizes/moves. Calls handle_stuff_while_moving_or_resizing(). (gdk_event_translate): Drop unused return_exposes parameter. Handle WM_XBUTTONDOWN and UP messages (buttons 4 and 5). On WM_SYSKEYUP, generate a key release event also for just the Alt key. On WM_MOUSELEAVE, generate a leave event of type GDK_NOTIFY_ANCESTOR (and not UNKNOWN) if the mouse left a top-level window, and left the app completely. On WM_ENTERSIZEMOVE, set a flag, and start an interval timer that calls resize_timer_proc() at regular intervals. On WM_EXITSIZEMOVE, kill the timer. On WM_WINDOWPOSCHANGED, generate a configure event if necessary, and dispatch the main loop (by calling handle_stuff_while_moving_or_resizing()). Fixes #99540, idea by Herman Bloggs. * gdk/win32/gdkmain-win32.c (_gdk_win32_message_to_string): Handle also wintab messages. * gdk/win32/gdkwindow-win32.c (gdk_window_set_skip_taskbar_hint): Instead of using WS_EX_TOOLWINDOW, implement by setting/unsetting the window's owner. Fixes #118093, reported by Maxime Romano.
2003-12-14 01:57:54 +00:00
static GdkWindow *owner = NULL;
GdkWindowAttr wa;
g_return_if_fail (GDK_IS_WINDOW (window));
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_skip_taskbar_hint: %p: %s\n",
GDK_WINDOW_HWND (window),
skips_taskbar ? "TRUE" : "FALSE"));
if (skips_taskbar)
Merge from stable: 2003-12-14 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/gdkdisplaymanager.c: Mark default_display static. * gdk/win32/gdkdnd-win32.c: Mark current_dest_drag static. * gdk/win32/gdkkeys-win32.c: Disable some overly verbose debugging output. * gdk/win32/gdkevents-win32.c: Clean up the debugging output from --gdk-debug=events. In general, output just one line of debugging output for each Windows message, plus one line for each GDK event generated. Indent all lines according to window procedure nesting level. (inner_window_procedure): Rename from real_window_procedure. (find_real_window_for_grabbed_mouse_event): Don't get misled when the point is in the non-client (decoration) area of the window returned by WindowFromPoint(). Return the root window in that case. (build_pointer_event_state): Test also MK_XBUTTON1 and MK_XBUTTON2 (buttons 4 and 5). (synthesize_enter_event): Track the mouse leaving the window in the event being generated, not the one mentioned in the Windows message. (propagate): Test for NULL parent earlier. Improves event generation from a grabbed pointer. Part of fix for #107320. (handle_stuff_while_moving_or_resizing): New function, to dispatch the main loop (once). (resize_timer_proc): New function, set to be called by an inerval timer during resizes/moves. Calls handle_stuff_while_moving_or_resizing(). (gdk_event_translate): Drop unused return_exposes parameter. Handle WM_XBUTTONDOWN and UP messages (buttons 4 and 5). On WM_SYSKEYUP, generate a key release event also for just the Alt key. On WM_MOUSELEAVE, generate a leave event of type GDK_NOTIFY_ANCESTOR (and not UNKNOWN) if the mouse left a top-level window, and left the app completely. On WM_ENTERSIZEMOVE, set a flag, and start an interval timer that calls resize_timer_proc() at regular intervals. On WM_EXITSIZEMOVE, kill the timer. On WM_WINDOWPOSCHANGED, generate a configure event if necessary, and dispatch the main loop (by calling handle_stuff_while_moving_or_resizing()). Fixes #99540, idea by Herman Bloggs. * gdk/win32/gdkmain-win32.c (_gdk_win32_message_to_string): Handle also wintab messages. * gdk/win32/gdkwindow-win32.c (gdk_window_set_skip_taskbar_hint): Instead of using WS_EX_TOOLWINDOW, implement by setting/unsetting the window's owner. Fixes #118093, reported by Maxime Romano.
2003-12-14 01:57:54 +00:00
{
if (owner == NULL)
{
wa.window_type = GDK_WINDOW_TEMP;
wa.wclass = GDK_INPUT_OUTPUT;
wa.width = wa.height = 1;
wa.event_mask = 0;
owner = gdk_window_new_internal (NULL, &wa, 0, TRUE);
Merge from stable: 2003-12-14 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/gdkdisplaymanager.c: Mark default_display static. * gdk/win32/gdkdnd-win32.c: Mark current_dest_drag static. * gdk/win32/gdkkeys-win32.c: Disable some overly verbose debugging output. * gdk/win32/gdkevents-win32.c: Clean up the debugging output from --gdk-debug=events. In general, output just one line of debugging output for each Windows message, plus one line for each GDK event generated. Indent all lines according to window procedure nesting level. (inner_window_procedure): Rename from real_window_procedure. (find_real_window_for_grabbed_mouse_event): Don't get misled when the point is in the non-client (decoration) area of the window returned by WindowFromPoint(). Return the root window in that case. (build_pointer_event_state): Test also MK_XBUTTON1 and MK_XBUTTON2 (buttons 4 and 5). (synthesize_enter_event): Track the mouse leaving the window in the event being generated, not the one mentioned in the Windows message. (propagate): Test for NULL parent earlier. Improves event generation from a grabbed pointer. Part of fix for #107320. (handle_stuff_while_moving_or_resizing): New function, to dispatch the main loop (once). (resize_timer_proc): New function, set to be called by an inerval timer during resizes/moves. Calls handle_stuff_while_moving_or_resizing(). (gdk_event_translate): Drop unused return_exposes parameter. Handle WM_XBUTTONDOWN and UP messages (buttons 4 and 5). On WM_SYSKEYUP, generate a key release event also for just the Alt key. On WM_MOUSELEAVE, generate a leave event of type GDK_NOTIFY_ANCESTOR (and not UNKNOWN) if the mouse left a top-level window, and left the app completely. On WM_ENTERSIZEMOVE, set a flag, and start an interval timer that calls resize_timer_proc() at regular intervals. On WM_EXITSIZEMOVE, kill the timer. On WM_WINDOWPOSCHANGED, generate a configure event if necessary, and dispatch the main loop (by calling handle_stuff_while_moving_or_resizing()). Fixes #99540, idea by Herman Bloggs. * gdk/win32/gdkmain-win32.c (_gdk_win32_message_to_string): Handle also wintab messages. * gdk/win32/gdkwindow-win32.c (gdk_window_set_skip_taskbar_hint): Instead of using WS_EX_TOOLWINDOW, implement by setting/unsetting the window's owner. Fixes #118093, reported by Maxime Romano.
2003-12-14 01:57:54 +00:00
}
SetWindowLong (GDK_WINDOW_HWND (window), GWL_HWNDPARENT,
(long) GDK_WINDOW_HWND (owner));
Merge from stable: 2003-12-14 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/gdkdisplaymanager.c: Mark default_display static. * gdk/win32/gdkdnd-win32.c: Mark current_dest_drag static. * gdk/win32/gdkkeys-win32.c: Disable some overly verbose debugging output. * gdk/win32/gdkevents-win32.c: Clean up the debugging output from --gdk-debug=events. In general, output just one line of debugging output for each Windows message, plus one line for each GDK event generated. Indent all lines according to window procedure nesting level. (inner_window_procedure): Rename from real_window_procedure. (find_real_window_for_grabbed_mouse_event): Don't get misled when the point is in the non-client (decoration) area of the window returned by WindowFromPoint(). Return the root window in that case. (build_pointer_event_state): Test also MK_XBUTTON1 and MK_XBUTTON2 (buttons 4 and 5). (synthesize_enter_event): Track the mouse leaving the window in the event being generated, not the one mentioned in the Windows message. (propagate): Test for NULL parent earlier. Improves event generation from a grabbed pointer. Part of fix for #107320. (handle_stuff_while_moving_or_resizing): New function, to dispatch the main loop (once). (resize_timer_proc): New function, set to be called by an inerval timer during resizes/moves. Calls handle_stuff_while_moving_or_resizing(). (gdk_event_translate): Drop unused return_exposes parameter. Handle WM_XBUTTONDOWN and UP messages (buttons 4 and 5). On WM_SYSKEYUP, generate a key release event also for just the Alt key. On WM_MOUSELEAVE, generate a leave event of type GDK_NOTIFY_ANCESTOR (and not UNKNOWN) if the mouse left a top-level window, and left the app completely. On WM_ENTERSIZEMOVE, set a flag, and start an interval timer that calls resize_timer_proc() at regular intervals. On WM_EXITSIZEMOVE, kill the timer. On WM_WINDOWPOSCHANGED, generate a configure event if necessary, and dispatch the main loop (by calling handle_stuff_while_moving_or_resizing()). Fixes #99540, idea by Herman Bloggs. * gdk/win32/gdkmain-win32.c (_gdk_win32_message_to_string): Handle also wintab messages. * gdk/win32/gdkwindow-win32.c (gdk_window_set_skip_taskbar_hint): Instead of using WS_EX_TOOLWINDOW, implement by setting/unsetting the window's owner. Fixes #118093, reported by Maxime Romano.
2003-12-14 01:57:54 +00:00
#if 0 /* Should we also turn off the minimize and maximize buttons? */
SetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE,
GetWindowLong (GDK_WINDOW_HWND (window), GWL_STYLE) & ~(WS_MINIMIZEBOX|WS_MAXIMIZEBOX|WS_SYSMENU));
SetWindowPos (GDK_WINDOW_HWND (window), NULL,
0, 0, 0, 0,
Merge from stable: 2003-12-14 Tor Lillqvist <tml@iki.fi> Merge from stable: * gdk/gdkdisplaymanager.c: Mark default_display static. * gdk/win32/gdkdnd-win32.c: Mark current_dest_drag static. * gdk/win32/gdkkeys-win32.c: Disable some overly verbose debugging output. * gdk/win32/gdkevents-win32.c: Clean up the debugging output from --gdk-debug=events. In general, output just one line of debugging output for each Windows message, plus one line for each GDK event generated. Indent all lines according to window procedure nesting level. (inner_window_procedure): Rename from real_window_procedure. (find_real_window_for_grabbed_mouse_event): Don't get misled when the point is in the non-client (decoration) area of the window returned by WindowFromPoint(). Return the root window in that case. (build_pointer_event_state): Test also MK_XBUTTON1 and MK_XBUTTON2 (buttons 4 and 5). (synthesize_enter_event): Track the mouse leaving the window in the event being generated, not the one mentioned in the Windows message. (propagate): Test for NULL parent earlier. Improves event generation from a grabbed pointer. Part of fix for #107320. (handle_stuff_while_moving_or_resizing): New function, to dispatch the main loop (once). (resize_timer_proc): New function, set to be called by an inerval timer during resizes/moves. Calls handle_stuff_while_moving_or_resizing(). (gdk_event_translate): Drop unused return_exposes parameter. Handle WM_XBUTTONDOWN and UP messages (buttons 4 and 5). On WM_SYSKEYUP, generate a key release event also for just the Alt key. On WM_MOUSELEAVE, generate a leave event of type GDK_NOTIFY_ANCESTOR (and not UNKNOWN) if the mouse left a top-level window, and left the app completely. On WM_ENTERSIZEMOVE, set a flag, and start an interval timer that calls resize_timer_proc() at regular intervals. On WM_EXITSIZEMOVE, kill the timer. On WM_WINDOWPOSCHANGED, generate a configure event if necessary, and dispatch the main loop (by calling handle_stuff_while_moving_or_resizing()). Fixes #99540, idea by Herman Bloggs. * gdk/win32/gdkmain-win32.c (_gdk_win32_message_to_string): Handle also wintab messages. * gdk/win32/gdkwindow-win32.c (gdk_window_set_skip_taskbar_hint): Instead of using WS_EX_TOOLWINDOW, implement by setting/unsetting the window's owner. Fixes #118093, reported by Maxime Romano.
2003-12-14 01:57:54 +00:00
SWP_FRAMECHANGED | SWP_NOACTIVATE | SWP_NOMOVE |
SWP_NOREPOSITION | SWP_NOSIZE | SWP_NOZORDER);
#endif
}
else
{
SetWindowLong (GDK_WINDOW_HWND (window), GWL_HWNDPARENT, 0);
}
}
void
gdk_window_set_skip_pager_hint (GdkWindow *window,
gboolean skips_pager)
{
g_return_if_fail (GDK_IS_WINDOW (window));
}
void
gdk_window_set_type_hint (GdkWindow *window,
GdkWindowTypeHint hint)
{
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at 2003-07-29 Tor Lillqvist <tml@iki.fi> Fix for #108007, #112402, #117042: There was confusion in gdk/win32 at various places whether a window position refers to the decoration position or the client area position. Also whether window size includes decorations or not. The correct interpretation apparently is that in GDK (like in X11), a top-level window position means the decoration's position, but size means the window's inner size (client area size). In the Win32 API, the window size usually includes the decorations, though. * gdk/win32/gdkevents-win32.c (decode_key_lparam): Move inside #ifdef G_ENABLE_DEBUG. (handle_configure_event): New function, generates GDK_CONFIGURE events from WM_SIZE and WM_MOVE messages. Even if no event is generated because of the event mask, still set the private position and size fields. Calculate position and size correctly. (gdk_event_translate): Call handle_configure_event(). * gdk/win32/gdkgeometry-win32.c: Cosmetics. * gdk/win32/gdkwindow-win32.c: Use GDI_CALL() and API_CALL() macros. Cosmetic debugging output changes. (SafeAdjustWindowRectEx): Remove. If an application wants to locate a window outside of the screen, it's not GDK's business to prevent it. And anyway, with multiple monitors, negative coordinates are perfectly normal. (gdk_window_new): Adjust the window size for decorations after _gdk_window_init_position() has done its job. (But the big window code currently is presumably broken on Win32 anyway.) (gdk_window_move): The position passed in is supposed to be that of the window border, so don't need to adjust for decorations. (gdk_window_resize, gdk_window_move_resize): Simplify somewhat.
2003-07-29 23:35:40 +00:00
GDK_NOTE (MISC, g_print ("gdk_window_set_type_hint: %p: %d\n",
GDK_WINDOW_HWND (window), hint));
((GdkWindowImplWin32 *)((GdkWindowObject *)window)->impl)->type_hint = hint;
update_style_bits (window);
}
GdkWindowTypeHint
gdk_window_get_type_hint (GdkWindow *window)
{
g_return_val_if_fail (GDK_IS_WINDOW (window), GDK_WINDOW_TYPE_HINT_NORMAL);
if (GDK_WINDOW_DESTROYED (window))
return GDK_WINDOW_TYPE_HINT_NORMAL;
return GDK_WINDOW_IMPL_WIN32 (((GdkWindowObject *) window)->impl)->type_hint;
}
void
gdk_window_shape_combine_region (GdkWindow *window,
GdkRegion *shape_region,
gint offset_x,
gint offset_y)
{
GdkWindowObject *private = (GdkWindowObject *)window;
g_return_if_fail (GDK_IS_WINDOW (window));
if (GDK_WINDOW_DESTROYED (window))
return;
if (!shape_region)
{
GDK_NOTE (MISC, g_print ("gdk_window_shape_combine_region: %p: none\n",
GDK_WINDOW_HWND (window)));
SetWindowRgn (GDK_WINDOW_HWND (window), NULL, TRUE);
private->shaped = FALSE;
}
else
{
HRGN hrgn;
hrgn = _gdk_win32_gdkregion_to_hrgn (shape_region, 0, 0);
GDK_NOTE (MISC, g_print ("gdk_window_shape_combine_region: %p: %p\n",
GDK_WINDOW_HWND (window),
hrgn));
do_shape_combine_region (window, hrgn, offset_x, offset_y);
private->shaped = TRUE;
}
}
void
gdk_window_input_shape_combine_region (GdkWindow *window,
GdkRegion *shape_region,
gint offset_x,
gint offset_y)
{
g_return_if_fail (GDK_IS_WINDOW (window));
/* Not yet implemented. See comment in
* gdk_window_input_shape_combine_mask().
*/
}
2001-03-31 19:33:49 +00:00
Changes multihead reorganizing code for win32 support, mostly from a patch Wed Jun 5 18:34:47 2002 Owen Taylor <otaylor@redhat.com> Changes multihead reorganizing code for win32 support, mostly from a patch by Hans Breuer. * gdk/gdkcolor.c gdk/x11/gdkcolor-x11.c gdk/gdkcursor.c gdk/x11/gdkcursor-x11.c gdk/gdkevents.c gdk/x11/gdkevents-x11.c gdk/gdkfont.c gdk/x11/gdkfont-x11.c gdk/gdkkeys.c gdk/x11/gdkkeys-x11.c gdk/gdkimage.c gdk/x11/gdkimage-x11.c gdk/gdkscreen.c gdk/x11/gdkmain-x11.c gdk/gdkdisplay.c gdk/gdkevents-x11.c gdk/gdkpango.c gdk/x11/gdkpango-x11.c gdk/gdkselection.c gdk/x11/gdkselection-x11.c gdk/gdkwindow.c gdk/x11/gdkwindow-x11.c gdk/gdkvisual.c gdk/x11/gdkvisual-x11.c: Move port-independent singlehead wrapper functions into port-independent part of GDK. (#80009) * gdk/win32/gdkcolor-win32.c gdk/win32/gdkcursor-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkfont-win32.c gdk/win32/gdkimage-win32.c gdk/win32/gdkkeys-win32.c gdk/win32/gdkmain-win32.c gdk/win32/gdkproperty-win32.c gdk/win32/gdkselection-win32.c gdk/win32/gkwindow-win32.c: Turn singlehead functions into "multihead" functions that ignore their GdkDisplay or GdkScreen arguments. * gdk/win32/gdkdrawable-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkinput-win32.c gdk/win32/gdkprivate-win32.h: Misc multihead-compatibility changes. * gtk/gtk.def gdk/gdk.def: Update for multihead functions. * gdk/gdkcolormap.h gdk/gdkvisual.h gdk/x11/gdkcolormap-x11.c gdk/x11/gdkvisual-x11.c: Remove the screen fields from the public parts of the colormap/visual structures, add accessors instead. * gdk/gdkpixbuf-render.c gdk/gdkpixmap.c gdk/gdkrgb.c gdk/x11/gdkcolormap-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkprivate-x11.h gtk/gtkgc.c gtk/gtkstyle.c gtk/gtkwidget.c: Use accessors to get the screen for colormaps, visuals; move the fields into the private structures for the x11 backend. * gdk/gdkdisplay.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/gdkscreen.[ch] gdk/x11/gdkscreen-x11.c: Remove virtualization of screen and display functions. (#79990, patch from Erwann Chenede) * gdk/win32/gdkdisplay-x11.c gdk/win32/gdkscreen-win32.c gdk/win32/{Makefile.am, makefile.msc, makefile.mingw}: New files containing stub implementations of Display, Screen functions. * gdk/x11/gdkscreen-x11.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/x11/gdkx.h: Clean up function exports and what headers they are in. (#79954) * gdk/x11/gdkx.h: Fix macro that was referring to a non-existant screen->screen_num. (In the patch for #79972, Erwann Chenede) * gdk/gdkscreen.c gdk/gdkwindow.c gdk/x11/gdkinternals.h gdk/x11/gdkscreen-x11.c: Fix gdk_screen_get_window_at_pointer() to use window hooks. (#79972, patch partly from Erwann Chenede) * gdk/x11/gdkdisplay-x11.c gdk/x11/gdkevents-x11.c: Fix some warnings.
2002-06-06 00:26:42 +00:00
GdkWindow *
gdk_window_lookup_for_display (GdkDisplay *display, GdkNativeWindow anid)
{
g_return_val_if_fail (display == _gdk_display, NULL);
Changes multihead reorganizing code for win32 support, mostly from a patch Wed Jun 5 18:34:47 2002 Owen Taylor <otaylor@redhat.com> Changes multihead reorganizing code for win32 support, mostly from a patch by Hans Breuer. * gdk/gdkcolor.c gdk/x11/gdkcolor-x11.c gdk/gdkcursor.c gdk/x11/gdkcursor-x11.c gdk/gdkevents.c gdk/x11/gdkevents-x11.c gdk/gdkfont.c gdk/x11/gdkfont-x11.c gdk/gdkkeys.c gdk/x11/gdkkeys-x11.c gdk/gdkimage.c gdk/x11/gdkimage-x11.c gdk/gdkscreen.c gdk/x11/gdkmain-x11.c gdk/gdkdisplay.c gdk/gdkevents-x11.c gdk/gdkpango.c gdk/x11/gdkpango-x11.c gdk/gdkselection.c gdk/x11/gdkselection-x11.c gdk/gdkwindow.c gdk/x11/gdkwindow-x11.c gdk/gdkvisual.c gdk/x11/gdkvisual-x11.c: Move port-independent singlehead wrapper functions into port-independent part of GDK. (#80009) * gdk/win32/gdkcolor-win32.c gdk/win32/gdkcursor-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkfont-win32.c gdk/win32/gdkimage-win32.c gdk/win32/gdkkeys-win32.c gdk/win32/gdkmain-win32.c gdk/win32/gdkproperty-win32.c gdk/win32/gdkselection-win32.c gdk/win32/gkwindow-win32.c: Turn singlehead functions into "multihead" functions that ignore their GdkDisplay or GdkScreen arguments. * gdk/win32/gdkdrawable-win32.c gdk/win32/gdkevents-win32.c gdk/win32/gdkinput-win32.c gdk/win32/gdkprivate-win32.h: Misc multihead-compatibility changes. * gtk/gtk.def gdk/gdk.def: Update for multihead functions. * gdk/gdkcolormap.h gdk/gdkvisual.h gdk/x11/gdkcolormap-x11.c gdk/x11/gdkvisual-x11.c: Remove the screen fields from the public parts of the colormap/visual structures, add accessors instead. * gdk/gdkpixbuf-render.c gdk/gdkpixmap.c gdk/gdkrgb.c gdk/x11/gdkcolormap-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkimage-x11.c gdk/x11/gdkprivate-x11.h gtk/gtkgc.c gtk/gtkstyle.c gtk/gtkwidget.c: Use accessors to get the screen for colormaps, visuals; move the fields into the private structures for the x11 backend. * gdk/gdkdisplay.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/gdkscreen.[ch] gdk/x11/gdkscreen-x11.c: Remove virtualization of screen and display functions. (#79990, patch from Erwann Chenede) * gdk/win32/gdkdisplay-x11.c gdk/win32/gdkscreen-win32.c gdk/win32/{Makefile.am, makefile.msc, makefile.mingw}: New files containing stub implementations of Display, Screen functions. * gdk/x11/gdkscreen-x11.[ch] gdk/x11/gdkdisplay-x11.[ch] gdk/x11/gdkx.h: Clean up function exports and what headers they are in. (#79954) * gdk/x11/gdkx.h: Fix macro that was referring to a non-existant screen->screen_num. (In the patch for #79972, Erwann Chenede) * gdk/gdkscreen.c gdk/gdkwindow.c gdk/x11/gdkinternals.h gdk/x11/gdkscreen-x11.c: Fix gdk_screen_get_window_at_pointer() to use window hooks. (#79972, patch partly from Erwann Chenede) * gdk/x11/gdkdisplay-x11.c gdk/x11/gdkevents-x11.c: Fix some warnings.
2002-06-06 00:26:42 +00:00
return gdk_window_lookup (anid);
}
Bug 143330, support update counter spec. Sun Jul 11 15:24:03 2004 Soeren Sandmann <sandmann@daimi.au.dk> Bug 143330, support update counter spec. * configure.in: Add check for the Sync extension * gdk/gdkwindow.h: gdk_window_enable_synchronized_configure() and gdk_window_configure_finished() * gtk/gtkwindow.c (gtk_window_move_resize): Call gdk_window_finish_configure(). * gtk/gtkwindow.c (gtk_window_realize): Automatically enable synchronized configures. * gdk/x11/gdkwindow-x11.h (struct _GdkToplevelX11): Store current and pending counter values. * gdk/x11/gdkwindow-x11.c (gdk_window_configure_finished): New function. * gdk/x11/gdkwindow-x11.c (gdk_window_enable_synchronized_configure): New function. * gdk/x11/gdkwindow-x11.c (gdk_toplevel_x11_free_contents): Delete update counter. * gdk/x11/gdkwindow-x11.c (set_wm_protocols): Advertise _NET_WM_SYNC_REQUEST when Sync is available * gdk/x11/gdkevents-x11.c (gdk_wm_protocols_filter): Handle _NET_WM_SYNC_REQUEST messages * gdk/x11/gdkevents-x11.c (gdk_event_translate): Save counter value for use by gdk_window_configure_finished() when receiving ConfigureNotifies. * gdk/x11/gdkdisplay-x11.h (struct _GdkDisplayX11): Add use_sync flag * gdk/x11/gdkdisplay-x11.c (gdk_display_open): Check if the XSync extension is available * gdk/x11/gdkdisplay-x11.c: Add _NET_WM_SYNC_REQUEST and _NET_WM_SYNC_REQUEST_COUNTER to list of supported atoms. * gdk/linux-fb/gdkwindow-fb.c, gdk/win32/gdkwindow-win32.c: Add stubs for enable_synchronized_configure() and configure_finished()
2004-07-11 13:26:57 +00:00
void
gdk_window_enable_synchronized_configure (GdkWindow *window)
{
g_return_if_fail (GDK_IS_WINDOW (window));
Bug 143330, support update counter spec. Sun Jul 11 15:24:03 2004 Soeren Sandmann <sandmann@daimi.au.dk> Bug 143330, support update counter spec. * configure.in: Add check for the Sync extension * gdk/gdkwindow.h: gdk_window_enable_synchronized_configure() and gdk_window_configure_finished() * gtk/gtkwindow.c (gtk_window_move_resize): Call gdk_window_finish_configure(). * gtk/gtkwindow.c (gtk_window_realize): Automatically enable synchronized configures. * gdk/x11/gdkwindow-x11.h (struct _GdkToplevelX11): Store current and pending counter values. * gdk/x11/gdkwindow-x11.c (gdk_window_configure_finished): New function. * gdk/x11/gdkwindow-x11.c (gdk_window_enable_synchronized_configure): New function. * gdk/x11/gdkwindow-x11.c (gdk_toplevel_x11_free_contents): Delete update counter. * gdk/x11/gdkwindow-x11.c (set_wm_protocols): Advertise _NET_WM_SYNC_REQUEST when Sync is available * gdk/x11/gdkevents-x11.c (gdk_wm_protocols_filter): Handle _NET_WM_SYNC_REQUEST messages * gdk/x11/gdkevents-x11.c (gdk_event_translate): Save counter value for use by gdk_window_configure_finished() when receiving ConfigureNotifies. * gdk/x11/gdkdisplay-x11.h (struct _GdkDisplayX11): Add use_sync flag * gdk/x11/gdkdisplay-x11.c (gdk_display_open): Check if the XSync extension is available * gdk/x11/gdkdisplay-x11.c: Add _NET_WM_SYNC_REQUEST and _NET_WM_SYNC_REQUEST_COUNTER to list of supported atoms. * gdk/linux-fb/gdkwindow-fb.c, gdk/win32/gdkwindow-win32.c: Add stubs for enable_synchronized_configure() and configure_finished()
2004-07-11 13:26:57 +00:00
}
void
gdk_window_configure_finished (GdkWindow *window)
Bug 143330, support update counter spec. Sun Jul 11 15:24:03 2004 Soeren Sandmann <sandmann@daimi.au.dk> Bug 143330, support update counter spec. * configure.in: Add check for the Sync extension * gdk/gdkwindow.h: gdk_window_enable_synchronized_configure() and gdk_window_configure_finished() * gtk/gtkwindow.c (gtk_window_move_resize): Call gdk_window_finish_configure(). * gtk/gtkwindow.c (gtk_window_realize): Automatically enable synchronized configures. * gdk/x11/gdkwindow-x11.h (struct _GdkToplevelX11): Store current and pending counter values. * gdk/x11/gdkwindow-x11.c (gdk_window_configure_finished): New function. * gdk/x11/gdkwindow-x11.c (gdk_window_enable_synchronized_configure): New function. * gdk/x11/gdkwindow-x11.c (gdk_toplevel_x11_free_contents): Delete update counter. * gdk/x11/gdkwindow-x11.c (set_wm_protocols): Advertise _NET_WM_SYNC_REQUEST when Sync is available * gdk/x11/gdkevents-x11.c (gdk_wm_protocols_filter): Handle _NET_WM_SYNC_REQUEST messages * gdk/x11/gdkevents-x11.c (gdk_event_translate): Save counter value for use by gdk_window_configure_finished() when receiving ConfigureNotifies. * gdk/x11/gdkdisplay-x11.h (struct _GdkDisplayX11): Add use_sync flag * gdk/x11/gdkdisplay-x11.c (gdk_display_open): Check if the XSync extension is available * gdk/x11/gdkdisplay-x11.c: Add _NET_WM_SYNC_REQUEST and _NET_WM_SYNC_REQUEST_COUNTER to list of supported atoms. * gdk/linux-fb/gdkwindow-fb.c, gdk/win32/gdkwindow-win32.c: Add stubs for enable_synchronized_configure() and configure_finished()
2004-07-11 13:26:57 +00:00
{
g_return_if_fail (GDK_IS_WINDOW (window));
Bug 143330, support update counter spec. Sun Jul 11 15:24:03 2004 Soeren Sandmann <sandmann@daimi.au.dk> Bug 143330, support update counter spec. * configure.in: Add check for the Sync extension * gdk/gdkwindow.h: gdk_window_enable_synchronized_configure() and gdk_window_configure_finished() * gtk/gtkwindow.c (gtk_window_move_resize): Call gdk_window_finish_configure(). * gtk/gtkwindow.c (gtk_window_realize): Automatically enable synchronized configures. * gdk/x11/gdkwindow-x11.h (struct _GdkToplevelX11): Store current and pending counter values. * gdk/x11/gdkwindow-x11.c (gdk_window_configure_finished): New function. * gdk/x11/gdkwindow-x11.c (gdk_window_enable_synchronized_configure): New function. * gdk/x11/gdkwindow-x11.c (gdk_toplevel_x11_free_contents): Delete update counter. * gdk/x11/gdkwindow-x11.c (set_wm_protocols): Advertise _NET_WM_SYNC_REQUEST when Sync is available * gdk/x11/gdkevents-x11.c (gdk_wm_protocols_filter): Handle _NET_WM_SYNC_REQUEST messages * gdk/x11/gdkevents-x11.c (gdk_event_translate): Save counter value for use by gdk_window_configure_finished() when receiving ConfigureNotifies. * gdk/x11/gdkdisplay-x11.h (struct _GdkDisplayX11): Add use_sync flag * gdk/x11/gdkdisplay-x11.c (gdk_display_open): Check if the XSync extension is available * gdk/x11/gdkdisplay-x11.c: Add _NET_WM_SYNC_REQUEST and _NET_WM_SYNC_REQUEST_COUNTER to list of supported atoms. * gdk/linux-fb/gdkwindow-fb.c, gdk/win32/gdkwindow-win32.c: Add stubs for enable_synchronized_configure() and configure_finished()
2004-07-11 13:26:57 +00:00
}
void
gdk_window_beep (GdkWindow *window)
{
gdk_display_beep (_gdk_display);
}
void
gdk_window_set_opacity (GdkWindow *window,
gdouble opacity)
{
LONG exstyle;
typedef BOOL (*PFN_SetLayeredWindowAttributes) (HWND, COLORREF, BYTE, DWORD);
PFN_SetLayeredWindowAttributes setLayeredWindowAttributes = NULL;
g_return_if_fail (GDK_IS_WINDOW (window));
g_return_if_fail (WINDOW_IS_TOPLEVEL (window));
if (GDK_WINDOW_DESTROYED (window))
return;
if (opacity < 0)
opacity = 0;
else if (opacity > 1)
opacity = 1;
exstyle = GetWindowLong (GDK_WINDOW_HWND (window), GWL_EXSTYLE);
if (!(exstyle & WS_EX_LAYERED))
API_CALL (SetWindowLong, (GDK_WINDOW_HWND (window),
GWL_EXSTYLE,
exstyle | WS_EX_LAYERED));
setLayeredWindowAttributes =
(PFN_SetLayeredWindowAttributes)GetProcAddress (GetModuleHandle ("user32.dll"), "SetLayeredWindowAttributes");
if (setLayeredWindowAttributes)
{
API_CALL (setLayeredWindowAttributes, (GDK_WINDOW_HWND (window),
0,
opacity * 0xff,
LWA_ALPHA));
}
}
void
_gdk_windowing_window_set_composited (GdkWindow *window, gboolean composited)
{
}