Commit Graph

365 Commits

Author SHA1 Message Date
okan
b46a5b0b56 Inline Xft draw and extents wrappers; too much abstraction. 2016-09-28 17:06:33 +00:00
okan
374d386034 Continue merging kb and mouse functions: fold
mousefunc_menu_{client,cmd,group} into the respective
kbfunc_menu_{client,cmd,group} functions; simply pass a flag down from
config denoting mouse action behaviour.
2016-09-22 14:36:03 +00:00
okan
489250a384 de-static client_inbound() 2016-09-20 19:11:19 +00:00
okan
9b943acf32 During init, query screen for _NET_ACTIVE_WINDOW and set that client as
active; while we already look at what's under the pointer, use this
information first, then look under the pointer (saving that round-trip).
This restores the active state to a client after restart even if the
pointer is not above it (and of course the pointer is not above another
client).
2016-09-16 14:32:02 +00:00
okan
403542f5b2 more client vs screen context differences 2015-11-12 21:28:03 +00:00
okan
db0907a53d Move kb pointer movement out of the kbfunc_client_moveresize since it's
got nothing to do with clients, thus doing flags work causes lots of
waste and almost useless jumpy pointer movements; while here, split out
move and resize since they share almost no code, just like mouse client
move/resize; factor out amount and factor. Still wonder why this is
here, but it works now.
2015-11-12 18:26:41 +00:00
okan
a7bbf5197e Partial revert of replacing screen_area() with region_find(); until a
fix for a regression is found; this bug has been around for a long time
it seems, but this change exposed it. Likely need to track clients in to
and out of regions.
2015-11-11 14:22:01 +00:00
okan
0abb5fd0f1 Start cleaning up name vs function differences; replace magic numbers. 2015-11-10 20:05:33 +00:00
okan
3fd7c4d329 Clean up unused defines. 2015-11-09 20:15:23 +00:00
okan
ffc7945b8d Extend region to include both view and work areas; switch to
region_find() which no longer needs to recalculate gap each time
a client (or menu) is created or altered. If no RandR, fall back
to display dimensions while building regions instead of during
execution.
2015-11-09 20:03:29 +00:00
okan
c10cac3b1e On execwm, we should properly release resources before exec'ing into a
new window manager; so allow CWM_EXEC_WM to assign new wm to wm_argv and
pass through cwm_status (now EXECWM) so that x_teardown() gets called
before exec'ing the new window manager.  Removes the need for a separate
x_restart() now, using new wm_argv; and consolidates errno for execvp.
2015-09-16 17:58:25 +00:00
okan
cc66816c13 Mechanical change: group->gc 2015-08-27 18:53:14 +00:00
okan
66a2bc1f37 Split out sticky mode checks and the restoring of a client's group and
_NET_WM_DESKTOP from the config-based auto-grouping; no (intentional)
behavior changes.  Needed for further work in cleaning up this area.
2015-08-25 18:29:10 +00:00
okan
ee0c4ddc88 Implement _NET_CLIENT_LIST_STACKING (from Thomas Admin), but
bottom-to-top order, as per spec (notified Thomas as well).
2015-08-24 15:42:57 +00:00
okan
6752f43126 Sort _NET_WM_STATE Atoms like the spec. 2015-08-24 14:56:10 +00:00
okan
b4a045b8c6 Add Xkb modifier to ignore mask; from Alexander Polakov. 2015-08-21 16:53:48 +00:00
okan
ba02c63913 Fix whitespace. 2015-08-21 16:52:37 +00:00
okan
1e8d4a7d04 Add client freeze extension to _NET_WM_STATE Atom, allowing flag to
persist. As usual with new Atoms, requires X restart.
2015-08-21 16:30:02 +00:00
okan
8529d9a1c6 introduce 'groupsearch' for group menu search; matches on either group
number/shortcut and/or name.
2015-07-12 14:31:47 +00:00
okan
6e095b76d0 Introduce a callback for cmd menu printing, special-casing 'lock' and
'term'.
2015-06-30 18:54:12 +00:00
okan
933ce723de Mechanical change from xinerama to region backed areas. 2015-06-26 17:17:46 +00:00
okan
44693cc984 collect 'gap' applications 2015-06-26 16:11:21 +00:00
okan
9660be8b0b Replace screen region info gathering with XRandR equivalent of Xinerama
queries (currently act on XRandR events anyway). Fall-back mode without
XRandR is still what X provides. This removes -lXinerama.
2015-06-26 15:21:58 +00:00
okan
6aef9ae536 typo in comment; from jsteinha [at] iutbayonne.univ-pau.fr 2015-06-09 13:02:15 +00:00
okan
5a1ff2b250 stash window dimensions 2015-06-08 15:11:29 +00:00
okan
055ed363c8 add a flag to indicate a menu list instead of overloading 'prompt' 2015-06-05 15:01:51 +00:00
okan
d6e84c8298 merge kbd and mouse grouptoggle 2015-05-21 00:37:04 +00:00
okan
2c1757a009 fill in mwm defines 2015-05-20 23:54:39 +00:00
kspillner
955b6eef83 Make window name and window class separate parameters to conf_autogroup.
No functional change, but will be used shortly.

ok okan@
2015-05-17 04:34:01 +00:00
okan
32d16969bc Introduce a xreallocarray and convert a few xcalloc instances that do
not require zero'ing.
2015-03-28 23:12:47 +00:00
okan
da47515de4 Simplify key/mb binding moving argtype into flags and dropping another
variable; removes the need to zero out struct binding, leaving a simple
malloc.
2015-03-26 21:41:43 +00:00
okan
c199720bcc Switch to limits.h; replace MAXPATHLEN and MAXHOSTNAMELEN with PATH_MAX
and HOST_NAME_MAX+1, respectively.

ok doug@
2015-01-19 14:54:16 +00:00
okan
72676125b1 make group_init work like other *_init's 2014-10-08 12:48:51 +00:00
okan
c9178afc38 these have nothing to do with 'sticky', but rather group membership; rename. 2014-09-27 19:04:32 +00:00
okan
9cd0504ba3 Update _NET_CURRENT_DESKTOP with the screen's group_active->num. 2014-09-23 14:25:08 +00:00
okan
b9aa350408 Move stuff that doesn't belong in group_init; while here, explicitly
initialize hideall and cycling.
2014-09-23 13:45:48 +00:00
okan
1d2cb82cf4 these client actions are just toggles; less confusing with better names 2014-09-17 18:41:44 +00:00
okan
05b4f3491a Use a similarly named check as sticky for hidden check in a group. 2014-09-17 16:32:53 +00:00
okan
8ffcf70d79 Implement EWMH _NET_WM_STATE_HIDDEN. 2014-09-17 16:00:44 +00:00
okan
d4ceb9f4e3 Introduce a check to see if a group holds only 'sticky' clients and use
this check to decide if a group is virtually empty.  Rationale: if a
group contains *only* 'sticky' clients, it should be skipped while
cycling through groups.  Apply similar logic to the group menu.

Based on an idea from phessler@, who also tested another version.
2014-09-17 14:31:37 +00:00
okan
3d7b8e2b10 fold in 'active' into 'flags' 2014-09-10 20:30:38 +00:00
okan
57ff8d4658 name the group client queue appropriately, like other queues 2014-09-08 21:15:14 +00:00
okan
0fefd96f3e Remove duplicate client queue (mruq); instead, remove and take the
global Clientq and place it inside screen_ctx since every client belongs
to a screen, then use the same per screen clientq to track stacking
order (the sole reason for mruq).
2014-09-08 20:11:22 +00:00
okan
898bea2d90 Now that a group knows its screen, only pass down the group_ctx. 2014-09-08 13:51:29 +00:00
okan
b5dd6e391c Add screen_ctx to group_ctx, and populate on init. 2014-09-07 20:57:26 +00:00
okan
ae092d4229 screen_fromroot -> screen_find 2014-09-07 17:38:38 +00:00
okan
63dc9657ad Get rid of a redundant array of groups per screen; we already have a
link to the group queue for each screen.
2014-09-07 17:27:20 +00:00
okan
70027868ed Move termpath and lockpath into cmdq; side effect is that 'lock' and
'term' now show up in the application menu.
2014-09-06 16:14:35 +00:00
okan
4abab66820 Rework group names: stash the group name inside the group_ctx as opposed
to having to manage an array outside in screen_ctx for group names and
shortcuts.  Simplifies (and moves bits for) reading, and constructing
data for, EWMH's _NET_DESKTOP_NAMES.
2014-09-06 16:08:58 +00:00
okan
de52ddf635 Merge group_menu into mousefunc_menu_group. 2014-09-01 18:04:58 +00:00
okan
ea1518cb5c Since the flatting out of sc->group_names is only for setting
NET_DESKTOP_NAMES, merge the helper into xu_ewmh_net_desktop_names,
where we actually set the property.
2014-08-25 14:31:22 +00:00
okan
749976d1d7 Implement _NET_WM_STATE_STICKY, bound to CM-s by default; allows any
client to 'stick' to all desktops (ewmh speak) or groups - this
currently has the same affect as setting a client's group to 'nogroup',
with the exception that the client can also be in a group, so when
un-sticking, the client will go back to its original group/desktop.
2014-08-25 12:49:19 +00:00
okan
4485c34f15 gc->hidden has never consistently kept track of a group's state;
group_show() and group_hide() are not the only ways a group can change
state - if all clients in a group are either hidden or unhidden, then
that group's state should change, as well as the various EWMH ways.
Instead of trying to keep track in a wide variety of places, simply
query the clients in a group before needing to take action based on the
group's state.  Solves long standing confusion of when a group is hidden
or not.
2014-08-24 15:37:45 +00:00
okan
b7cd8019f2 Purely mechanical; unify 'num', 'no' and 'shortcut'. 2014-08-20 15:15:29 +00:00
okan
7fd50277e7 Pull highstack from group_ctx (and useless calculations of); in the one
place that we use highstack, replace that usage with a local variable
(for now until stacking is done properly).
2014-08-19 12:47:51 +00:00
okan
c4aace036f Get rid of nhidden in group_ctx; it actually never reported correctly
since nhidden wasn't incremented nor decremeted in all the right places,
thus confusing matters.  We don't need to carry a count around, so just
use a local variable in the one place we need one to supply
XRestackWindows().
2014-08-18 13:57:57 +00:00
okan
875128a0cb Make sure we cycle through existing client's group membership and set
the hidden state on the group; important to know group hidden state
after a re-exec or restart of cwm (as easily seen in group menu).
2014-02-08 02:49:30 +00:00
okan
16de3733be Introduce a region queue and replace screen's XineramaScreenInfo; we
still use Xinerama to populate the regions per screen, but will switch
at a more appropriate time.
2014-02-03 21:07:47 +00:00
okan
ea2abae9f3 Move redundant bits from screen_init (while dealing with existing
clients) directly into client_init, performing the X roundtrip only
once. With the previous change in maprequest, this moves decision making
into one place for creating new clients.
2014-02-03 20:20:39 +00:00
okan
03847f1c2e No need to store screen colormap and visual; rather just use the X
macros in the one place they are needed.
2014-02-02 16:29:04 +00:00
okan
4ce9b05438 The menu already limits entries with MENU_MAXENTRY, so don't bother
holding a command name limit as well.
2014-02-02 15:46:05 +00:00
okan
4de6836838 Switch ignoreq to winname struct since it's basically the same thing;
removes limit on name matching.
2014-01-30 22:17:22 +00:00
okan
9f5aa96cc7 Move cwm_status around to avoid header fu, and remove extraneous signal
header from xevents.c; noticed by Christian Neukirchen.
2014-01-30 15:41:11 +00:00
okan
06d9319772 Remove duplicate mouse functions and instead use the kbfunc ones. No
user visable changes at this point, but they'll merge at an appropriate
time.
2014-01-30 14:40:21 +00:00
okan
d5979e3ab6 Minimize trivial differences between a few kb and mb functions. 2014-01-29 22:30:00 +00:00
okan
78ff1f748c Merge keybinding and mousebinding queues into using the same merged
struct, binding; they were essentially the same accept for what was
'pressed', keysym or button.
2014-01-29 21:13:52 +00:00
okan
2ffb124daa Check command name/path for truncation and provide user feedback during
config parse (and use conf_cmd_add to populate defaults); based on a
discussion with Tiago Cunha.  While this looks ugly, there are likely
some other changes here to come.
2014-01-29 18:34:22 +00:00
okan
0590f4c1fc Check ignore windowname for truncation and provide user feedback during
config parse; based on a discussion with Tiago Cunha.
2014-01-28 20:22:21 +00:00
okan
3b8ad25f38 Somewhat streamline event loop/restart/quit handling; most notable
change allows a restart to trigger proper teardown first, even though
teardown is not (yet) complete.

After some discussion with oga@nicotinebsd.org regarding a more
complicated version/idea.
2014-01-22 21:48:27 +00:00
okan
b92e846e56 Sprinkle a few more const; from Tiago Cunha. 2014-01-21 15:42:44 +00:00
okan
57bbf0fd90 merge KBFLAG_NEEDCLIENT and MOUSEBIND_CTX_*; brings kbfunc and mousefunc
bits even closer.
2014-01-20 23:03:51 +00:00
okan
42e2d07b2b Use argument to pass down flags for mousefunc cyclegroup; removes
rcyclegroup wrapper need - now similar to kbfunc.
2014-01-20 22:31:53 +00:00
okan
9a540a7adb constify and rename some confusing variables around cmdq. 2014-01-20 21:34:32 +00:00
okan
7e2aa0c80b - remove redundant range check for buttons in conf_bind_mouse.
- make conf_bind_kbd return error on non-matches to match what
   conf_bind_mouse does.
 - rename some variables while here for clarity.
 - constify bind and cmd.

from Tiago Cunha.
2014-01-20 19:06:04 +00:00
okan
83b0cbf116 Add a function that adds an entry to a menuq, normalizing a common code
path; from Tiago Cunha.
2014-01-20 18:58:03 +00:00
okan
72ccee09ff use consistent types 2014-01-03 15:29:06 +00:00
okan
24e4f6a4b6 bring mousefunc closer to kbfunc 2014-01-02 21:30:20 +00:00
okan
424dc78f8e gc keycode in keybinding since we only deal with keysym now 2014-01-02 21:17:23 +00:00
okan
468c12c5da Implement support for EWMH's _NET_WM_STATE_FULLSCREEN hint.
Since we already have a form of 'maximize', we need to differentiate
between 'maximize' and the new 'fullscreen' mode.  The 'maximize' mode
will continue to honor gap but now *retains* the border, matching the
'vert/horz maximize' behaviour.  The new 'fullscreen' mode supports and
follows the _NET_WM_STATE_FULLSCREEN hint, allowing the client perform
additional window modifications; in this mode, cwm(1) will *ignore* gap,
remove borders and freeze(move/resize) the client.  Additionally,
'fullscreen' mode will remember various combinations of previous states.

* default keybinding changes: CM-f 'fullscreen', CM-m 'maximize' (re-map
  as desired).

Positive feedback from a few, testing and ok sthen@
2013-12-16 19:02:17 +00:00
okan
4104232e02 Teach screen_find_xinerama() to apply gap only when told to do so;
adjust callers.  Needed for an upcoming feature.
2013-12-13 22:39:13 +00:00
okan
1d68a42ec0 Add support for XUrgency and matching _NET_WM_STATE_DEMANDS_ATTENTION
ewmh hint; urgencyborder is configurable.  The urgency flag will stick,
even while on a client in a non-viewable group, until the client
receives focus (where the border is reset).  Initial diff from Thomas
Adam with some changes/enhancements from me.
2013-12-13 14:40:52 +00:00
okan
e5da774880 ICCCM explicitly states that server time (CurrentTime) should *not* be
used for focus events, but rather the timestamp of the generated event.
Track the last event timestamp and send it down for a WM_TAKE_FOCUS
ClientMessage.  I suspect we should do this for clients that don't
announce this Atom as well, though the raciness gets us into a bind.

Solves focus order issue since WM_TAKE_FOCUS; fix verified by sthen@

ok sthen@
2013-12-12 20:15:07 +00:00
okan
3d5c10942f Add client wrapper for XWMHints to support XA_WM_HINTS in PropertyNotify
events; based off a diff from Thomas Adam.
2013-12-11 15:41:11 +00:00
okan
5da88fd68f Remove extra work and simplify client state handling. 2013-12-11 14:16:09 +00:00
okan
33eb08de92 Stash Class and WM Hints in client_ctx 2013-12-11 14:09:21 +00:00
okan
a287dfc793 like gap, make snapdist per screen 2013-11-27 18:34:34 +00:00
okan
57ba45de2c Rewrite active/inactive client handling in client_setactive();
client_leave() served no real purpose, likewise no reason to handle
LeaveNotify events since an EnterNotify will process the next active
client (and we don't have anything important to process anyway), so
xev_handle_leavenotify() goes as well.  Allows a simplification of
client_mtf() and client_cycle_leave() for clarity.  While here, unify a
few client_current() checks.

No intended behaviour change.
2013-11-27 00:01:23 +00:00
okan
632f11c555 Alter the r1.35 of event.c race fix. Remove the forward looking event
queue check (removing the need for a server grab/ungrab) - if the client
is going away, let it fall all the way through to a DestroyNotify event.
There's no longer a need for us to manually destroy a client ourselves
(removing yet another server grab/ungrab).  Instead, when the
UnmapNotify event is synthetic, simply set the state to Withdrawn (as
per ICCCM), else Iconic (in our case 'hidden').

Verified with test case from the 2009 race which was the original reason
for r1.35 of event.c.
2013-11-12 21:25:00 +00:00
okan
e8fcfc29d4 stash WMProtocols in flags 2013-11-08 17:35:12 +00:00
okan
d21ac16a79 The only reason we need to keep XSizeHints in our client_ctx is for
flags, so just add one to hints; consolidates sizehints and shrinks.
Additionally don't abuse PSize for XGetWMNormalHints() failures.
2013-11-02 19:13:56 +00:00
okan
95e47782f7 re-add support for WM_TAKE_FOCUS, and additionally this time only call
XSetInputFocus() for clients that have the InputHint; latter fix
discovered by Valery Masiutsin with a PoC patch - solves keyboard input
focus loss for java apps.
2013-11-01 14:07:19 +00:00
okan
a8efa775ea A client_delete should behave differently depending on whether the
triggering event was unmap (with pending destroy) log destroy; we only
need to grab/ungrab the server lock, set WithdrawnState and
XRemoveFromSaveSet clients coming from an unmap event - doing so for
clients coming from destroy are already gone, hence we generate errors.
2013-10-25 19:46:16 +00:00
okan
f22f30d08a wrap key ungrab like btn ungrab, for it'll be used again 2013-10-20 01:55:32 +00:00
okan
d5f92c3e42 Using xu_btn_ungrab() buttons during client_leave doesn't work (error
BadValue) when the modifier is already AnyModifier .  Instead alter
xu_btn_ungrab() to ungrab AnyButton/AnyModifier and call it only when a
client is coming into focus in client_setactive(), instead of iterating
over ignore mods - matches how we handle key grabs.
2013-10-19 19:39:34 +00:00
okan
78f4db5140 unify type; no change 2013-10-03 13:51:57 +00:00
okan
cb56ed0c01 max -> nitems 2013-07-16 14:04:44 +00:00
okan
db709659c2 shuffle down protoypes 2013-07-15 23:53:19 +00:00
okan
1ca1aa39ab simplify atom handling; allows us to limit to one round-trip to server
for gathering Atoms.
2013-07-15 14:50:44 +00:00
okan
0cd6dbc0fc since the root and event window are the same in the case of a button
event on the screen's root window, there's no need to pass down the
entire XButtonEvent event, at least to group_menu(), the only callback
which takes an argument at this point; instead use the already populated
screen.
2013-07-09 01:24:49 +00:00
okan
c2e750f350 add support for mouse based group {,r}cycle; from Rodrigo Mosconi. not
bound by default.
2013-07-08 18:39:20 +00:00