2002-11-03 Havoc Pennington <hp@pobox.com>
* src/display.c (meta_display_grab): remove XSync calls from here
(meta_display_ungrab): remove XSync from here, but put in
an XFlush to be sure we get the ungrab sent.
* src/util.c (meta_topic): track sync count here
* src/errors.c: move sync count out of here
Throughout: error spew on all XSync() calls
* src/run-metacity.sh: don't set METACITY_DEBUG
2002-11-03 Havoc Pennington <hp@pobox.com>
* src/window.c (meta_window_new): use window-props.h stuff for a
couple of properties
(implement_showing): fix printf string
* src/xprops.c (meta_prop_free_values): new function
* src/window-props.h, src/window-props.c: start moving code that
handles loading window properties into this file.
2002-10-28 Havoc Pennington <hp@pobox.com>
* src/window.c, src/display.c: store the window menu on the
display and blow it away when a window closes, so we don't
get funny stuck menus. Patch from Martin Garton #87514
2002-10-25 Havoc Pennington <hp@pobox.com>
Add "busy cursor on app startup" support, conditionally - works
only if libstartup-notification is found, and in practice requires
a GTK patch that's not in yet.
* src/screen.c: monitor startup events and set busy cursor if
appropriate
* src/display.c (meta_display_open): create SnDisplay
* configure.in: check for startup notification,
and add the cute "configure summary" at the end
2002-10-21 Havoc Pennington <hp@redhat.com>
Optimizations for managing new windows (do not all take effect if
METACITY_DEBUG=1). Bug #96404
* src/keybindings.c (meta_change_keygrab): use error trap nesting
and conditionalize on meta_is_verbose() to avoid a ton of XSync
* src/display.c (meta_change_button_grab): ditto
Throughout: move to new error trap setup to save on XSync calls,
new setup is:
* src/errors.c (meta_error_trap_push_with_return): new function,
an error trap that needs to care about return value and thus
sync even if an outer trap still exists
(meta_error_trap_pop_with_return): new function
(meta_error_trap_pop): add "last_request_was_roundtrip"
argument allowing us to avoid XSync() if we just did
a GetProperty or whatever.
* src/util.c (meta_warning): flush the warning file descriptor
* src/Makefile.am (INCLUDES): define G_LOG_DOMAIN
2002-10-16 Havoc Pennington <hp@redhat.com>
* src/workspace.c: workspaces are all per-screen now, fix
accordingly
* src/core.c: fix multihead workspace stuff
* src/keybindings.c: multihead-rama
* src/screen.c (meta_screen_show_desktop): new functions to
replace display equivalents
* src/display.c (meta_display_get_workspace_by_screen_index): get
rid of this
(meta_display_get_workspace_by_index): get rid of this
(event_callback): handle _NET_SHOWING_DESKTOP message per-screen
* src/screen.c (meta_screen_get_workspace_by_index): new function
* src/screen.h (struct _MetaScreen): move workspace list, and
showing_desktop flag, to be per-screen
* src/window.c (window_query_root_pointer): return whether pointer
is on window's screen
(meta_window_handle_mouse_grab_op_event): don't use coordinates
from other screens when updating a window operation on the current
screen. I can't believe no one has reported this...
2002-10-14 Federico Mena Quintero <federico@ximian.com>
* src/display.c (event_callback): Ignore EnterNotify events when
the detail field is set to NotifyInferior. Fixes#95747.
2002-10-11 Havoc Pennington <hp@redhat.com>
* src/display.c (event_callback): don't raise window on button 2
click, only on button 1 and button 3.
* src/frames.c (meta_frames_button_press_event): lower on button 2
press on frame
* src/core.c (meta_core_user_lower): new function
2002-10-07 Havoc Pennington <hp@redhat.com>
Add a modifier key preference for the Alt+click stuff.
Can be set to "disabled" as well.
* src/run-metacity.sh: load .Xmodmap in the Xnest if it exists
* src/display.c (meta_display_ungrab_window_buttons): ungrab
AnyModifier in case the modifier changed since we grabbed
(meta_display_open): rearrange code to use meta_display_close() to
mop up when we can't find any screens, avoiding the need to
keep the bail-out code in sync with meta_display_close.
* src/keybindings.c (devirtualize_modifiers): move this function
to a public place in display.c
* src/metacity.schemas.in: add setting for the modifier key
to use for Alt+left/middle/right click.
* src/prefs.c (update_binding): add a missing newline to a warning
(meta_prefs_get_mouse_button_mods): new function
* src/ui.c (meta_ui_parse_modifier): new function
2002-10-04 Havoc Pennington <hp@redhat.com>
* src/display.c (event_callback): do XRRUpdateConfiguration()
if we have RandR extension, else poke in Xlib's screen struct to
update the screen size.
* configure.in: fix a bogus overwrite of cppflags,
add a check for RandR extension
2002-10-01 Havoc Pennington <hp@pobox.com>
* src/screen.c (meta_screen_resize_func): make it static
* src/stack.c (get_standalone_layer): put above/below windows
in an appropriate layer.
* src/screen.c (set_supported_hint): say we support above/below
* src/display.h (struct _MetaDisplay): add _NET_WM_STATE_ABOVE,
_NET_WM_STATE_BELOW atoms
* src/window.c (meta_window_client_message): handle above/below
state messages
(set_net_wm_state): handler above/below state
(update_net_wm_state): handle above/below states
2002-09-28 Havoc Pennington <hp@pobox.com>
Patch from Keith Packard to handle root window resizes.
* src/screen.c (reload_xinerama_infos): factor out Xinerama code
(meta_screen_resize): implement this, to be called from display.c
on screen resize
* src/display.c (event_callback): handle ConfigureNotify on root
windows
2002-09-27 Havoc Pennington <hp@redhat.com>
Try to handle Solaris Xinerama, all coded blind, someone
on Solaris will need to debug the typos.
* src/display.c: updates for Solaris Xinerama
* src/screen.c: updates for Solaris Xinerama
* configure.in: make Xinerama check more complicated to catch
Solaris Xinerama
2002-09-23 Havoc Pennington <hp@pobox.com>
* src/main.c (main): re-enable the log handler, maybe it will
break something, I don't remember why I turned it off.
* src/display.c: s/_NET_SHOW_DESKTOP/_NET_SHOWING_DESKTOP/ which
is what's in the spec
2002-09-22 Havoc Pennington <hp@pobox.com>
* src/window.c (recalc_window_features): small reordering of
code
* src/display.c (meta_spew_event): more spew for MapNotify,
UnmapNotify
* src/window.c (recalc_window_features): spew more stuff
* src/display.c (meta_spew_event): spew override_redirect field of
ConfigureNotify
2002-09-03 Havoc Pennington <hp@pobox.com>
* src/display.c (meta_display_get_tab_current): new function
* src/keybindings.c (do_choose_window): apply modified patch from
JeyaSudha to still display tab popup if only one window is on the
desktop.
2002-08-25 Havoc Pennington <hp@pobox.com>
* src/display.c (meta_display_open): _NET_SUPPORTING_WM_CHECK is
supposed to have type WINDOW not CARDINAL. reported by
Ben Jansens
2002-08-20 Havoc Pennington <hp@redhat.com>
* src/frames.c (get_control): if in the title rect check for y
<= TOP_RESIZE_HEIGHT
* src/display.c (meta_spew_event): put x/y coordinates in spew for
enter/leave notify
* src/frames.c (meta_frames_motion_notify_event): move cursor
changing from here to update_prelit_control so it happens on enter
notify as well
(get_control): change test "y < TOP_RESIZE_HEIGHT" to
"y <= TOP_RESIZE_HEIGHT"
* src/Makefile.am (EXTRA_DIST): include .in files in EXTRA_DIST
2002-08-15 Havoc Pennington <hp@pobox.com>
* src/window.c (recalc_window_features): leave has_fullscreen_func
set to TRUE if the window is screen sized and undecorated, even if
the window isn't resizable. idea from Christian - Manny Calavera -
Neumair
* src/keybindings.c (handle_toggle_fullscreen)
(handle_toggle_maximize): these disabled fullscreen/maximize if
the window wasn't resizable, should have used has_fullscreen_func
has_maximize_func instead.
2002-08-15 Havoc Pennington <hp@pobox.com>
* src/keybindings.c: implement raise/lower
* src/metacity.schemas.in: add raise/lower
* src/prefs.c: add "raise" and "lower" prefs to keybindings
* src/display.c (meta_display_set_grab_op_cursor): assert that
the screen arg is non-NULL in appropriate cases
2002-08-14 Jayaraj Rajappan <jayaraj.rajappan@wipro.com>
* src/display.c (meta_display_set_grab_op_cursor):
In XGrabPointer, set the confine_to argument to the root window
of the screen the window is on.
* src/display.h: add screen argument.
* src/window.c (meta_window_update_resize_grab_op):
pass screen argument as NULL.
2002-08-12 Havoc Pennington <hp@redhat.com>
* src/stack.c (compute_layer): window is in fullscreen layer if
any member of its group is fullscreen
* src/window.c (meta_window_unmake_fullscreen): update layer for
whole window group
(meta_window_make_fullscreen): ditto
* src/util.c (meta_unsigned_long_hash): move hash/equal funcs for
Window in here.
* src/group.c: track window groups so we can do stuff with them.
2002-08-10 Havoc Pennington <hp@pobox.com>
* src/window.c (update_resize): track time to avoid sending a
deluge of move/resize requests, suggestion from
xavier.bestel@free.fr bug #86830. Not really sure if this will
make a difference or not. We'll see I guess.
(update_move): do same on move though it seems less important
here.
* src/display.h (struct _MetaDisplay): store the
last time we sent a move/resize event.
2002-08-08 Craig Black <blackc@speakeasy.net>
Patch to provide extra cues to the user when using
window menu move and resize, #85724.
* src/common.h: add new cursors
* src/display.c: (grab_op_is_mouse)
(meta_display_create_x_cursor), (xcursor_for_op),
(meta_display_set_grab_op_cursor),
(meta_display_begin_grab_op):
The keyboard move and resize grab ops now also use the mouse.
Allow the grab cursor to be changed during the grab op.
Hold onto the initial grab position in case of reset.
* src/display.h: save the initial grab position
* src/keybindings.c: (process_keyboard_move_grab),
(process_keyboard_resize_grab), (handle_begin_move),
(handle_begin_resize):
The keyboard move and resize grab ops now also use the mouse.
* src/window.c: (meta_window_client_message), (menu_callback),
(update_move), (update_resize),
(meta_window_handle_mouse_grab_op_event), (warp_pointer),
(meta_window_warp_pointer), (meta_window_begin_grab_op),
(meta_window_update_resize_grab_op):
When moving or resizing a window use the last grab position
in computing change increment.
Provide support for warping the mouse pointer.
* src/window.h: new warp pointer and grab op helper functions
2002-08-07 Craig Black <blackc@speakeasy.net>
* src/display.c: (meta_display_open): clear expected focus window
on open
* src/display.h: add expected_focus_window field
* src/window.c: (meta_window_make_fullscreen),
(meta_window_unmake_fullscreen): change meta_window_update_layer()
to meta_stack_update_layer() so build works again.
(meta_window_free), (meta_window_make_fullscreen),
(meta_window_focus), (meta_window_notify_focus): keep track of
expected focus window after sending WM_TAKE_FOCUS event,
previously if a UnmapNotify event arrived before the FocusIn event
we would lose focus, fixes#84564.
2002-07-28 Havoc Pennington <hp@pobox.com>
* src/window.c (meta_window_shade): disable animation when shading
windows, just doesn't really convey the idea anyway.
* src/effects.c: Move to using a shaped window instead of
IncludeInferiors to do the animations, looks a lot better
because we don't have to grab the server.
* src/window.c (meta_window_change_workspace): remove bogus
assertion that was causing a crash
(meta_window_new): auto-fullscreen huge undecorated windows.
* src/keybindings.c (switch_to_workspace): use
meta_window_change_workspace() to avoid same bug in cut-and-paste
code from there
2002-08-04 Havoc Pennington <hp@redhat.com>
* src/frames.c (meta_frames_button_press_event): raise/focus
on click, even if the click was on the client area
(this makes Alt+button1 raise windows again, yay)
* src/stack.c (compute_layer): put panels in the DOCK layer always
(keep them on top of other windows). Still sloppy-focus raised
with respect to other docks.
* configure.in: remove -Wshadow for now as GTK headers make all
kinds of noise with it.
2002-08-04 Havoc Pennington <hp@redhat.com>
* src/theme.c (free_menu_ops): use MetaMenuIconType not button
type for the size of the menu ops array
(meta_theme_define_int_constant): return TRUE on success (how the
heck did this ever work?)
(meta_theme_define_float_constant): return TRUE on success
(meta_frame_style_validate): allow the "positional" buttons to
be omitted for now.
* src/testgradient.c (render_multi): don't define N_COLORS twice
* src/theme-viewer.c (run_theme_benchmark): don't define
ITERATIONS twice
* src/theme.c (button_rect): handle new button types
(meta_button_type_to_string): update
(meta_button_type_from_string): update
* src/theme.h (enum): add button types for the 6 possible button
positions. No way to reposition buttons still but this will allow
themes to go ahead and support doing so.
2002-08-03 Craig Black <blackc@speakeasy.net>
* src/display.c: (event_callback): Have ButtonPress and UnmapNotify
events account for a null grab window, fixes#87896
2002-08-01 Mark McLoughlin <mark@skynet.ie>
Implements support for _NET_WM_ALLOWED_ACTIONS.
Fixes#84282.
* src/display.[ch]: (meta_display_open): add
_NET_WM_ALLOWED_ACTIONS atoms.
* src/screen.c: (set_supported_hint): set them
as being supported.
* src/window.c:
(set_allowed_actions_hint): impl setting
_NET_WM_ALLOWED_ACTIONS.
(recalc_window_features): use it here, but only
if things have changed.
2002-07-20 Havoc Pennington <hp@pobox.com>
* src/display.c (meta_display_open): grab display across managing
each screen; XGetInputFocus() on startup.
2002-07-19 Havoc Pennington <hp@pobox.com>
* src/window.c (meta_window_configure_request): disable configure
requests during a user move/resize operation, mostly a workaround
for stoopid apps.
2002-07-23 Havoc Pennington <hp@redhat.com>
* src/keybindings.c (meta_display_process_key_event): handle
NULL screen from screen_for_xwindow
* src/display.c (meta_display_screen_for_xwindow): put an error
trap around the XGetWindowAttributes(), should fix the popular
"closing a window results in a crash" bug.
* src/util.c (print_backtrace): support optional backtrace
feature using gnu libc backtrace() call
2002-07-06 Havoc Pennington <hp@pobox.com>
Apply blackc@speakeasy.net patch, bug #83940, to do
mini-workspaces similar to the pager, when switching
spaces.
* src/window.c (update_net_wm_state): actually fill in
wm_state_skip_taskbar, wm_state_skip_pager flags
* src/tabpopup.c: support drawing a mini-workspace similar to the
one the pager draws.
* src/stack.c (meta_stack_list_windows): new function to list
the windows in stacking order
* src/screen.c (meta_screen_ensure_workspace_popup): don't pass in
the ugly default app icon for workspaces
* src/display.c (event_callback): fix from blackc@speakeasy.net
to avoid dereferencing a NULL grab window.
2002-07-06 Havoc Pennington <hp@pobox.com>
* src/display.c (meta_display_open): put _NET_DESKTOP_NAMES in the
array of atom names, so desktop names might work and we don't read
uninitialized memory.
* src/main.c (main): add VERSION/timestamp verbose message.
* src/keybindings.c: implement cycle_windows cycle_panels
* src/metacity.schemas.in: add the cycle_windows cycle_panels
keybindings
* src/prefs.h (META_KEYBINDING_FOCUS_PREVIOUS): replace
FOCUS_PREVIOUS key binding with CYCLE_WINDOWS and CYCLE_PANELS
(not good names really, but I don't have ideas).
* src/common.h: add a grab op for alt+esc window cycling
2002-07-01 Mark McLoughlin <mark@skynet.ie>
Fix broken workspace switching from my previous commit.
* src/display.c: (meta_display_begin_grab_op): don't
leak a pointer grab if we fail to grab the keyboard.
* src/keybindings.c: (meta_screen_grab_keys): check
screen->all_keys_grabbed.
(meta_screen_grab_all_keys): regrab our standard
bindings if we fail.
(handle_workspace_switch): revert to our previous
behaviour of using the last focused window to do
the grab upon. Only use the RootWindow if there
isn't anything else to use.
* src/screen.c: (meta_screen_new): initialise
all_keys_grabbed.
2002-06-26 Mark McLoughlin <mark@skynet.ie>
Fixes not being able to tab out of a
workspace which contains no windows.
* src/core.c: (meta_core_begin_grab_op): upd
for meta_display_begin_grab_op change.
(meta_core_get_grab_frame): allow for
grab_window == NULL.
* src/display.[ch]:
(meta_display_screen_for_xwindow): implement.
(meta_display_begin_grab_op): grab on the root window
if window == NULL.
(meta_display_end_grab_op): use grab_screen instead of
grab_window.
* src/keybindings.c:
(grab_all_keys_and_keyboard): split out from
meta_window_grab_all_keys.
(ungrab_all_keys_and_keyboard): split out from
meta_window_ungrab_all_keys.
(meta_screen_grab_all_keys), (meta_screen_ungrab_all_keys):
implement grabbing and ungrabbing on the root window.
(meta_display_process_key_event): if window == NULL,
check the event is from the same screen and process. Only
happens with workspace switching.
(process_workspace_switch_grab): kill window param and
don't use grab_window.
(handle_tab_forward), (handle_begin_move): upd for
meta_display_begin_grab_op change.
(handle_workspace_switch): remove brokeness. Always do
the grab op on the root window.
* src/keybindings.h: add meta_screen_{un}grab_all_keys.
* src/window.c: (meta_window_client_message), (menu_callback):
update for meta_display_begin_grab_op change.
2002-06-25 Mark McLoughlin <mark@skynet.ie>
* src/display.[ch]: (meta_display_open):
src/screen.c: (set_supported_hint), (set_work_area_hint):
Its _NET_WORKAREA, not _NET_WM_WORKAREA silly :-)
2002-06-25 Mark McLoughlin <mark@skynet.ie>
* src/screen.[ch]:
(update_num_workspaces), recalc workarea hint when
new workspaces created. Fixes bug that workarea
not calculated until first non-dock window is
mapped.
(set_work_area_hint), (set_work_area_idle_func),
(meta_screen_queue_workarea_recalc): move all this
stuff from workspace.c.
* src/workspace.c: (meta_workspace_invalidate_work_area):
use meta_screen_queue_workarea_recalc.
2002-06-22 Havoc Pennington <hp@pobox.com>
* src/workspace.c (set_number_of_spaces_hint): do nothing if
screen is being unmanaged, we don't want to blow away state,
we want to remember it for the next window manager.
2002-06-22 Havoc Pennington <hp@pobox.com>
* src/workspace.c (meta_screen_ensure_workspace_popup): rename
from meta_workspace_ensure_tab_popup, and use workspace->name
instead of a hardcoded name
2002-06-22 Havoc Pennington <hp@pobox.com>
* src/xprops.c (meta_prop_get_utf8_list): new utility function
* src/display.c (meta_display_open): _NET_DESKTOP_NAMES atom
(event_callback): update workspace names when the property changes
* src/screen.c (set_supported_hint): "support" _NET_DESKTOP_NAMES
(nothing to do really)
2002-06-09 Havoc Pennington <hp@pobox.com>
* src/frames.c (meta_frames_button_press_event): don't raise/focus
the window if minimize/close are clicked, patch from Gaute
Lindkvist #75460
2002-06-08 Havoc Pennington <hp@pobox.com>
Cleanups to workspace popup patch. Space before all parens
in a couple places.
* src/prefs.c (meta_prefs_get_keybinding_action): fix brace
indentation, and use while instead of for loop. Take a "mask"
argument to avoid ambiguity issues.
* src/keybindings.c (handle_workspace_switch): rename from
handle_workspace_forward since there's no directionality here
(handle_workspace_switch): add a FIXME about how screwed it is
that we need a window in order to do our grab. Should be able to
grab on a dummy window like no_focus_window or the root window.
(process_workspace_switch_grab): rename from tab_grab for clarity,
no tab involved here.
* src/common.h (enum): have only one grab op for all workspace
switching directions, instead of one for each.
2002-06-08 Havoc Pennington <hp@pobox.com>
Apply big patch from blackc@speakeasy.net adding a popup window
to the Ctrl+Alt+arrows shortcuts. #83940
2002-06-08 Havoc Pennington <hp@pobox.com>
* src/screen.c (meta_screen_new): select key press/release on the
display->no_focus_window, another attempted fix for not getting
keybindings when no window is focused. Still doesn't seem to work
though. I don't get what's going wrong.
(meta_create_offscreen_window): new function, used instead of
XCreateSimpleWindow so we get override redirect offscreen windows.
2002-06-08 Havoc Pennington <hp@pobox.com>
* src/display.c (meta_display_open): set net_supporting_wm_check
in addition to win_supporting_wm_check, patch from
JeyaSudha for #83365
* src/screen.c (set_wm_check_hint): remove setting
win_supporting_wm_check on leader window here, done already in
display.c
2002-06-08 Havoc Pennington <hp@pobox.com>
* src/xprops.c (meta_prop_get_utf8_string): don't die on bad atom
name
* src/display.c (meta_display_close): don't unmanage windows here,
do it in screen_free and then closing the display unmanages
windows as a side effect of unmanaging the screen
(meta_display_unmanage_screen): new function
(process_selection_clear, process_selection_request): handle
selection stuff
(meta_spew_event): don't crash on client message containing
invalid atom
(meta_spew_event): don't crash on property notify with invalid
atom
* src/main.c (main): add --replace option to replace existing
window manager.
* src/screen.c: implement holding manager selection.
* src/display.c (meta_display_open): add new selection-related
atoms.
