mirror/cwm
1
0
Fork 0
mirror of https://github.com/leahneukirchen/cwm.git synced 2023-08-10 21:13:12 +03:00
Code Issues Projects Releases Wiki Activity
729 Commits 4 Branches 18 Tags 2.3 MiB
6b8b856217
Commit Graph

196 Commits

Author SHA1 Message Date
okan
c1bc6d37b3 quick keyboard focus fix for clients that neither populate wmhints nor wmprotocols, like rdesktop; focus needs to be re-visited 2013-11-05 00:55:42 +00:00
okan
b5915eb989 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
b8f53666bd x/y from XSizeHints are obsolete (and have been for a long time), so 
instead use x/y from XWindowAttributes when USPosition|PPosition are
set.
 2013-11-01 21:54:20 +00:00
okan
06eb13dfd9 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
6e17b41f56 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
9ff288baef revert 1.138 (WM_TAKE_FOCUS) for now 2013-10-20 01:35:47 +00:00
okan
a2013ee9dd 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
083a023f2c I believe we redraw the border in too many cases; likely a leftover from 
the cc->pwin days - don't redraw on every unhide, resize, and mouse
move/resize Expose event (note, all Expose events trigger a redraw
anyway).

Tested with some finicky X apps I could think of, though I'm sure others
will find more - if so, and they 'lose' the border, please report!
 2013-10-19 18:59:22 +00:00
okan
a70b2d81af For clients that support WM_TAKE_FOCUS in their WM_PROTOCOLS property, send 
a ClientMessage event.
 2013-10-19 00:24:54 +00:00
okan
6ac51b8e41 unify type; no change 2013-10-03 13:51:57 +00:00
okan
4ef96a2772 collapse lines 2013-07-15 23:51:59 +00:00
okan
4119b5c065 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
6059073a5f move synthetic responses and have client_msg only work with WM_PROTOCOLS, 
since that's all ClientMessageEvent is for anyway.
 2013-06-10 21:37:30 +00:00
okan
be1a7a3f5c alter conf_grab(_kbd) to first ungrab AnyKey/AnyModifier, then proceed 
to grab keys in keybindingq.  we don't need to ungrab/grab on every
addition to the queue, just once with a complete keybindingq; simplify
grabbing keys per screen (during init) and during a MappingNotify.

while here, change conf_grab_{kbd,mouse} to require only a Window.
 2013-05-23 16:52:39 +00:00
okan
284af4ca0c add support for _NET_WM_STATE_MAXIMIZED_{HORZ,VERT}; from Alexander Polakov. 
while I'm unsure of the final look of _NET_WM_STATE, there's no reason
to delay this support.
 2013-05-20 21:13:58 +00:00
okan
dac00a232f - switch border colors to Xft 
- merge border/menu color structures/functions since they now both use Xft
- switch xu_xorcolor to operating on XftColor instead of just
  XRenderColor (basically adding pixel)
- if color name allocation fails, revert back to default (this, along
  with font validation should occur during config parse, but we don't
  have screens setup yet - likely to change at some point)
 2013-05-19 23:09:59 +00:00
okan
01bfdd7b42 use XGetWMProtocols and simplify WM_PROTOCOL handling 2013-05-19 17:02:04 +00:00
okan
2b926784ba new -> init 2013-05-11 22:01:07 +00:00
okan
1db7cc2a18 remove group in client_delete directly. 2013-05-06 16:03:11 +00:00
okan
006a29e617 mechanical xu_{get,set}state -> xu_{get,set}_wm_state change 2013-04-17 13:57:06 +00:00
okan
302690624e slightly rework WM_STATE set/get to make it less ambigious; will be more clear 
on what needs to change to make it right in the end.
 2013-04-17 13:52:20 +00:00
okan
16ed8bf8e4 only a window is required to set WM_STATE. also un-confuse xu_ptr_getpos 
by using 'win' instead of 'rootwin' so as not to imply only the root
window is queried, rather any window.
 2013-04-14 16:13:17 +00:00
okan
f3dfc4968c we handle WM_STATE here, so remove misleading comment. 2013-04-12 20:54:27 +00:00
okan
ef1b78f464 we already set window state in client_hide or client_unhide right before, so 
there's no need to do it again.
 2013-04-12 20:45:57 +00:00
okan
670207f79b zap stray whitespace 2013-04-05 17:07:25 +00:00
okan
c5a76f1d5b - no reason to breakout ClassHints work, so fold into client_{new,delete}, 
but keep _MOTIF_WM_HINTS separate.
- simplify fetching app/class hint.
- fix _MOTIF_WM_HINTS Atom type.
 2013-04-03 20:22:55 +00:00
okan
ec4474a33a move XUngrabServer to the end of client_new() to avoid races where clients, 
such as those using sdl, attempt to manage the clients themselves when the
clients aren't fully ready.  other wm's grab the xserver during the whole
client setup process, so match.

behavior found by jsg.
 2013-04-03 19:20:50 +00:00
okan
3a7596968b add per-group vert/horiz tiling support; introduces 2 new bind commands, 
'vtile' and 'htile'; from Alexander Polakov.
 2013-01-08 15:16:04 +00:00
okan
bf9d981597 teach screen_find_xinerama() about gap and adjust (simplify) callers; 
menu becomes gap-aware for free.
 2013-01-08 04:12:51 +00:00
okan
e7b85cfb2f fix menu/client placement in panning setups; XineramaQueryScreens gives 
us the width of the psuedo screen, but here we need the edge instead
(xmax/ymax); just re-use w/h here for now.
 2013-01-07 21:53:23 +00:00
okan
c426254da5 use cc->bwidth in client_vmax since we've yet to reset it (matches 
client_hmax); from Jan Stary
 2013-01-06 01:01:26 +00:00
okan
6d123cd96d rename {h,v}max functions for consistency; from Jan Stary 2013-01-04 16:30:03 +00:00
okan
698530155d get rid of struct color 2013-01-04 16:23:04 +00:00
okan
2ac65bd288 re-work client_snapcalc() so it takes client and edge dimensions with 
snapdist; allows for simplier snap calculations.

required for an upcoming diff for honoring gap.
 2013-01-02 21:37:21 +00:00
okan
ce8ef02ed2 have screen_find_xinerama() return struct geom *reliably* instead of 
XineramaScreenInfo; simplifies goop around the callers.
 2013-01-02 16:26:34 +00:00
okan
e492ed8e41 more variable consistency 2013-01-02 02:19:20 +00:00
okan
2d28b71a1b info->xine, for consistent variable names 2013-01-02 02:02:08 +00:00
okan
6cb334e503 replace emptystring with strdup here as well so we know we are free'ing 
a malloc'd variable everytime; from Tiago Cunha
 2013-01-01 14:26:29 +00:00
okan
0957fbbee2 make num of groups no longer off-by-one; from Alexander Polakov 
note that a re-exec of cwm will not rewrite the group number atom of
*existing* clients, so they will remain off-by-one until each client has
its atom updated, or of course a restart of X.
 2013-01-01 14:19:56 +00:00
okan
fe44d1dfce cmap -> colormap in client_ctx 2012-12-17 18:35:26 +00:00
okan
82e8ec4245 replace client highlight with a client flag 2012-12-17 17:48:57 +00:00
okan
8e67d1389c make client_mtf static 2012-12-17 14:58:46 +00:00
okan
d651c1cc3b current_client() returns _curcc, so use it where appropriate 2012-12-17 14:26:29 +00:00
okan
28224ff830 sort 2012-11-09 03:52:02 +00:00
okan
76b0874b4c get rid of the xfree() wrapper around free(); from Tiago Cunha. 2012-11-07 20:34:39 +00:00
okan
86524e21dd extend client_resize so that it can know when to reset max flags and 
bwidth; this allows a client to be resized from a max state, which now
gets treated like a non-max'd client.  based on a diff that does part of
this in a different way from Alexander Polakov.
 2012-09-09 19:47:47 +00:00
okan
3f956098d9 unify various max routines to match client placement and partially menu ones 
when calculating screen dimensions.
 2012-07-16 01:53:14 +00:00
okan
e42d1c7f01 move the HasXinerama check into screen_find_xinerama and simplify logic; 
moves closer to logic in menu code.
 2012-07-16 01:36:30 +00:00
okan
0e8815dfb0 re-use geom struct in client_ctx (saved)geometry. 2012-07-13 17:01:04 +00:00
okan
186a78ff1e convert xmax/ymax uses to view geometry. 2012-07-13 15:21:35 +00:00
okan
8c7964468f fix missing atom (WM_PROTOCOLS) wrt new style handing; without this, the 
window manager didn't know if a client supported CLIENT_PROTO_DELETE and
thus used a hammer, XKillClient.

behaviour reported by Tim van der Molen.
 2012-07-04 18:00:13 +00:00
okan
f98e123bfc re-implement atom handing; makes for a normalized and more consistent 
separation between cwm and ewmh.  seen by a few.
 2012-07-03 13:49:03 +00:00
okan
ac82403a51 cycle through other common cycling modifiers; based on a diff from 
Alexander Polakov.

ok sthen@
 2012-05-16 01:09:17 +00:00
okan
57766d8606 altpersist->cycling to be more clear that this is for cycling mod 
persistence; based on a diff from Alexander Polakov, in preparation for
looping through other potential cycling modifiers.
 2012-05-13 15:17:13 +00:00
okan
fb4936d0f7 knf, some from a diff from Tiago Cunha. 2012-05-13 15:15:54 +00:00
okan
796b32123d add WM_TRANSIENT_FOR event support: moves dialogs, toolbars and such to 
the group of the main application window; based on a diff from Alexander
Polakov with CLIENT_IGNORE flag suggestion from oga@.

ok oga@
 2011-09-13 08:41:57 +00:00
okan
d85b3adc0c restore a comment and add another for clarity. 2011-09-05 07:37:55 +00:00
oga
44d8b1d3ac Make flavours of maximisation additive. 
i.e. horiz-max + vertmax = full maximisation.  full - horiz = vertmax.
etc.

Martynas wrote something like this once, so I did okan, this version
seems to finally deal with the corner cases.

ok okan@.
 2011-09-04 16:59:31 +00:00
okan
b852a73a60 split off window hints from geometry so we don't need to carry them all 
around when dealing with {,h,v}max.  same idea from oga.
 2011-09-03 09:42:33 +00:00
okan
142a36a0c0 Add {r,}cycleingroup to cycle through clients belonging to the same 
group as the active client (as opposed to all unhidden clients); from
Alexander Polakov, with a tiny tweak requested by oga.

ok oga@
 2011-09-03 09:20:58 +00:00
okan
c585dfbf32 correct spelling; from Alexander Polakov. 2011-07-14 11:39:53 +00:00
okan
34ba6cdbcc introduce a new config option to snap to the screen edge. 'snapdist' 
keyword taken from a diff from Sviatoslav Chagaev to do the same thing,
but implemented in a completely way (based on some very old code from
mk@).  default set to 0, so no behavior change.

ok oga@ (who would also like to take it further...)
 2011-06-24 06:06:24 +00:00
okan
b230e611be re-do various bits of {,h,v}max for simplicity and less flag handling. 
as a bonus, flipping between various combinations of vmax+hmax provides
a more predictable behavior.

commitski oga@
 2011-06-24 06:01:47 +00:00
okan
f6fcd0eb5d alter a few function returns and prototypes; found by lint. 
ok oga@
 2011-06-24 05:40:09 +00:00
okan
c8cd94882d tag and comment cleanup; ok oga@ 2011-05-11 13:53:51 +00:00
okan
43450c8fd7 introduce a new 'freeze' flag (CMS-f by default) which may be applied to 
any window, after which all move/resize requests will be ignored,
essentially freezing the window in place.

there's a possibility to merge this with the 'ignore' concept, pending
on how ignore+freeze should behave (really more ewmh stuff), but punting
for now since ponies are on the line.

requested and tested by thib at k2k11 with ponies, unicorns and rainbows.

'save the unicorns' todd@, ok oga@
 2011-05-07 17:15:37 +00:00
okan
8f88cd474a if virt/horz un-maximizing, restore only the un-maximized axis 
coordinates; allows moving a client and not restoring the maximized axis
coordinates.

picked from a larger martynas diff from ages ago.

ok oga@
 2011-03-22 10:56:08 +00:00
okan
3ff303658c while this piece deals with client argv's in a seemingly inefficient way 
(noticed by Tim Peniket), since we don't ever do anything with cliarg,
remove the hunk and cliarg completely.

ok oga@
 2011-03-22 10:52:37 +00:00
okan
a12fb346a2 remove XXX and move a configure event out of the event handler functions. 
reminded by a similiar diff from Thomas Pfaff.

ok oga@
 2011-03-22 10:49:46 +00:00
okan
64e62989d0 if we are saving the pointer location for a specific client, make sure 
that if it is not already inbounds, put it in the default location,
which happens to be inbounds.

behavior noticed by Thomas Pfaff while maximizing and un-maximizing a
window, leaving the pointer behind and causing client_cycle() to be a
bit lost.

ok oga@
 2011-02-13 20:09:57 +00:00
okan
72e6d2d0a9 replace XFetchName() with something more intelligent which attempts to 
use the appropriate netwm Atom first, as well as deal with utf8.

slightly different incarnation tested by sthen@ and ajacoutot@ - thanks!

ok oga@
 2010-05-22 22:10:31 +00:00
okan
b12e6fc5cd clean up a few xu_* functions to just accept what they need (Window). 
ok oga@
 2010-04-11 16:51:26 +00:00
stsp
c4a97053cd Honour program-specified window position in size hints when placing windows. 
Makes XMMS windows appear as expected.

ok okan, oga
 2010-02-02 19:28:46 +00:00
okan
e88bda0df5 - allow per-screen gap; not (yet) user configurable. 
- teach _NET_WORKAREA about gap.

ok oga@
 2010-01-27 03:04:50 +00:00
okan
ccb207a8a8 pull all non-X11 headers from calmwm.h and place them only where they 
are required.

encourged to go all the way by oga@
 2009-12-15 04:10:42 +00:00
okan
b27c3c22b5 pull these headers only into files that need them. 
ok oga@
 2009-12-15 03:34:34 +00:00
okan
b6b7d273d4 merge the 2 common header files; specific includes to be pulled out as 
separate commits.

ok oga@
 2009-12-15 03:24:36 +00:00
oga
134e777cf0 finish unfucking the screen_ctx handling. 
remove screen_current() it was utterly bogus when nscreens > 1.

pass a fake client_ctx in the case where there's no client and the
kbfunc or mousefunc doesn't need a real one, it just contains the
current screen, modify these functions so that they pass down the screen
context to their callees.

make groups per screen, it's the only way it makes sense in this regard.

ok okan@.
 2009-12-10 17:16:51 +00:00
oga
728d2a40dd support _NET_CLIENT_LIST. 
the x property api doesn't let you remove one entry from an X property
array, so client_remove is kinda expensive, but there's no real way
around that..

ok okan@
 2009-12-07 22:46:15 +00:00
oga
3c60d854db Implement _NET_ACTIVE_WINDOW. for now just the informational hint is 
supported, the client message to change this will be supported when all
informational hints are working.

ok okan@
 2009-12-07 22:21:59 +00:00
tobias
42bf29fb20 style (whitespaces) 
ok oga, okan
 2009-11-28 17:52:12 +00:00
oga
85d8697676 With Xinerama enabled, the borders of a maximized window will show up on 
all adjacent screens. This patch hides the borders while the window is
maximized.

From Bertrand Janin; thanks!

ok okan@
 2009-09-25 15:57:49 +00:00
okan
7295c51155 unroll the CCTOSC macro; from Thomas Pfaff 
ok oga@
 2009-08-27 01:38:08 +00:00
oga
8fd4ff1c7c Instead of messing around everytime we do a resize, just clamp the 
resize increments to a minimum of one, and use it unconditionally.

"you've convinced me, do it!" okan@
 2009-08-25 01:32:40 +00:00
okan
fcb2684db1 unroll macro; ok oga@ 2009-08-25 01:26:09 +00:00
oga
e2b1cb98c1 Add a keybinding to allow horizontal maximisation of a window (CMS-enter). 
based on a diff by Thomas Pfaff; thanks!

ok okan@
 2009-08-24 23:54:41 +00:00
okan
028a1778db bring together gathering, calculating and applying of size hints; 
additionally, respect aspect ratio hints.

ok oga@
 2009-08-24 23:49:04 +00:00
oga
09d88f4a18 Instead of having a function that just calls TAILQ_INIT on a global, use 
TAILQ_HEAD_INITIALIZER() and drop the function.

ok okan@
 2009-08-24 21:22:48 +00:00
oga
e7f0d63413 instead of checking for flags in the client context, then removing them. 
e.g.:

if (flags & flags_we_care_about)
	flags &= ~(flags_we_care_about);

just whack the flags unconditionally, it's simpler.

okan@ agrees.
 2009-08-24 17:04:39 +00:00
okan
f44862be9c static local functions and data; almost identical diff from Thomas Pfaff 
ok oga@
 2009-06-26 12:21:58 +00:00
okan
b13d592c57 tidy up startup/init routines 
ok oga@
 2009-06-23 21:52:38 +00:00
okan
bcc0f73bb6 compact a bit by condensing a few if-else's; from Thomas Pfaff 
"go on then" oga@
 2009-06-20 00:55:41 +00:00
okan
58d12134b1 unroll XCALLOC/XMALLOC macros; since we use xcalloc/xmalloc all over the 
place anyway, this makes things a bit more consistent; from Thomas Pfaff

ok oga@
 2009-06-20 00:22:39 +00:00
okan
18c7d89c98 spacing 2009-06-20 00:19:56 +00:00
okan
61f841ea58 re-factor parts of mouse move/resize bit to shrink and make more 
readable; no behavior change

ok oga@
 2009-06-17 12:45:01 +00:00
okan
d1b84c5415 re-order a bit for readability. 
"if it makes you happy" oga@
 2009-05-30 00:30:17 +00:00
okan
2c29a1de65 nuke the leading underscore notation for local static functions - there 
are far better ways to know.

"go for it" oga@
 2009-05-18 00:23:35 +00:00
okan
4d5dc5d9ea a long time coming - re-work the way we deal with colors: since we're 
using Xft(3), use it to select the font color as well instead of trying
to build one; properly allocate and free colors at-will, e.g. we now
have configurable colors.

feedback and ok's todd@ and oga@
 2009-05-17 23:40:57 +00:00
sthen
6df7cba24e redraw the border when unhiding a client window. fixes the situation where 
you change to a different group and the mouse isn't over a window; previously
the border of the previously active window was highlighted but it didn't
actually have focus.

reads ok to oga@.
 2009-05-17 16:51:43 +00:00
okan
9203c7e8ca passing a null pointer to free() is valid; sprinkle a few free->xfree. 
ok oga@
 2009-01-23 20:47:45 +00:00