972 Commits

Author SHA1 Message Date
op
9eb763ab87 cwm: fix a semi-transparency border issue with some applications
When running with a compositor the border of some applications (firefox,
chromium, zathura...) is not rendered correctly.  Initializing the highest
significant bits of the color fixes it.

diff from Julien Blanchard (julien at typed-hole [dot] org); ok okan@
2023-03-22 08:27:36 +00:00
4a6128d5e4 Fix overlapping menu items as reported by kn@ and probably others at
some point; with op@ and Walter Alejandro Iglesias helping along the
way.  Consistently use font ascent+descent instead of an arbitrary font
height+1 for individual menu item rectangles and placement.

ok kn@ op@
2022-10-15 16:06:07 +00:00
4e73ce533c cycling fix: when no client is active, warp pointer to last active;
from Walter Alejandro Iglesias.
2022-02-27 14:59:55 +00:00
496bcc879d whitespace 2022-02-26 15:19:18 +00:00
5e5221d82d Fix spelling of some unused MWM hints; from Sean C. Farley.
While here, flesh out the rest of the MWM hints.
2022-02-26 15:03:42 +00:00
op
7c22b36a23 Add group-last command that shows only the previously active group; ok okan 2022-01-27 18:45:10 +00:00
81a08ddb89 Allow bare numbers for key and mouse bindings; taken from similar
support in other parse.y's; from Leon Fischer <lfischer@airmail.cc>.
2021-12-24 16:00:47 +00:00
a9eeb04606 sync parse.y changes from base; ok naddy@
original from naddy@:
> Don't declare variables as "unsigned char *" that are passed to
> functions that take "char *" arguments.  Where such chars are
> assigned to int or passed to ctype functions, explicitly cast them
> to unsigned char.
>
> For OpenBSD's clang, -Wpointer-sign has been disabled by default,
> but when the parse.y code was built elsewhere, the compiler would
> complain.
>
> With help from millert@
> ok benno@ deraadt@
2021-11-22 00:51:54 +00:00
055b84f4d4 Do not attempt to grab keys without a keycode; this incidentally allows
XF86 keys support.

found and fix by Luis Henriques <henrix@camandro.org>
2021-11-19 19:13:14 +00:00
kn
d46f34f01e Keep pointer within window on maximize/fullscreen toggle
Spawn a window, maximize it in any way, move the cursor to a window border
that is not on the screen's edge and unmaximize again:  While the window
goes back the cursor stays at the screen's edge, i.e.  focus is lost to the
underlaying window.

Moving, resizing, tiling or snapping windows in any way always moves the
cursor along iff needed, e.g. using MS-[hjkl] to move a small window from
the center to the edge keeps the cursor within window borders -- no matter
what you do with the keyboard, focus stays on that window.

Make CM-f, CM-m, CM-equal and CMS-equal (default bindings) for toggling
full-screen mode, maximization, vertical maximization and horizontal
maximization of the current window drag the cursor along if needed as well.

OK okan kmos dv
2021-04-22 10:02:55 +00:00
91c05f9403 Fixed memory leak in xu_get_strprop.
If a client calls XSetTextProperty for a window to clear all its
properties, then allocated memory within libX11 is not freed.

OK okan@
2020-04-25 20:07:28 +00:00
3ebe04ee8e Prevent out of boundary write with configuration files in which too many
quoted arguments are stored for other window managers.

The quotation handling happens within the while loop without checking if
the "end" limit has been already reached. If this happens, the final
NULL assignment leads to an out of boundary write on stack.

OK okan@
2020-04-16 17:12:49 +00:00
6407eb9bc1 Allow configuring a percentage window size of the master window during
htile/vtile actions. From Uwe Werler, with a few manpage tweaks.
2020-04-16 13:32:35 +00:00
6afdd483c7 zap stray tabs 2020-03-24 14:48:29 +00:00
6c20772841 Instead of using _NET_ACTIVE_WINDOW on restart, use the pointer location
to determine what client to set active. Reduces a round trip for every
window.
2020-03-24 14:47:29 +00:00
0a7d8cc5c4 Add support for SIGINT/SIGTERM. 2020-03-23 20:14:27 +00:00
tim
6c7b8261df Simplify conditional construct.
OK okan@
2020-03-20 18:50:08 +00:00
49f839e194 Trim event_mask to those that the root window actually needs. 2020-03-20 15:16:31 +00:00
207b71ef1c No need to lookup current client early; move to right before it is
needed.
2020-03-20 12:13:20 +00:00
tim
3b9b98c024 Recommit 1.259, but now with TAILQ_FOREACH_SAFE.
From and OK okan@

Original commit message:

Plug two memory leaks. Also get rid of a variable that is no longer
necessary.

OK okan@
2020-03-16 17:50:44 +00:00
tim
b9213d0a02 Revert previous. Causes a crash as reported by Tom Murphy. 2020-03-14 16:11:09 +00:00
tim
146fa08e4d Simplify list markup.
OK okan@ schwarze@
2020-03-13 20:50:07 +00:00
tim
d8c7d87737 Plug two memory leaks. Also get rid of a variable that is no longer
necessary.

OK okan@
2020-03-13 20:49:13 +00:00
3d1a8028c2 Remove ColormaskChange from event-mask since there's no event handler. 2020-02-28 13:38:35 +00:00
96275a835d Unrelated style fixes, consistency changes and sorting, appropriate
dosage/removal of wrappers, simplification of name queue, client cycle joins
other kb/mb bound functions.
2020-02-27 14:56:39 +00:00
7a88b2bdb2 Allow the 'empty' group clients to be window-{h,v}tile'd.
Behaviour (or lack there of) noticed by Raf Czlonka.
2020-02-07 18:53:41 +00:00
d3410dd10d Map ('5') and allow mod5mask (altgr) as a modifier.
From Artturi Alm (though changed from 'm' to '5')
2020-02-03 16:38:02 +00:00
83de84b7f8 add, then use, xvasprintf, checking for appropriate return. 2020-01-22 19:58:35 +00:00
2fc191f978 Ensure the pointer stays within client bounds after a window 'snap' (to edge).
reported by Stefan Hagen.
2020-01-21 15:50:03 +00:00
781af9c998 Because cwm warps the pointer during a client move (to stay within the client),
there's a window of time where an expose or enternotify event will get
generated for a lower client; use a hammer and drain events after keyboard
move/resize, until such a time that cwm doesn't warp the pointer.  Behavior
noticed by kn.

ok kn@
2019-08-13 18:45:38 +00:00
6131e36f8a Fix regression from r1.107 (lost a return); kettenis@ pointed out the high
potential for a use-after-free (true!) where kn@ ran into the regression using
an app that actually exercised the XGrabPointer() failure path.
2019-08-12 00:52:36 +00:00
kn
fdb841c3b0 Add application section
Link what is described as "applications" here to how they are actually
defined in cwmrc(5).

While here, call the configuration file what it is instead of reusing
the default path (already mentioned in the FILES section).

OK okan
2019-07-09 21:38:44 +00:00
kn
40000724cc command uses execvp(3) not execve(2)
util.c:u_exec() has been doing so since import.
2019-07-02 23:37:47 +00:00
tim
25b699b582 Plug a memory leak in log_debug(); OK okan@ 2019-04-29 19:03:20 +00:00
tim
3f1caab46a Handle _NET_WM_NAME changes.
This fixes the problem where cwm's window menu wouldn't show Firefox's current
window title if it contains non-ASCII characters.

OK okan@
2019-04-29 19:02:21 +00:00
9a7528f5b9 Check the atom type on propertynotify before iterating. 2019-03-11 15:25:46 +00:00
e55c0d48fa use screen_find() for xrandr crtc changes 2019-03-10 22:53:11 +00:00
5bc2098c6f Find the managed screen from the parent window for client_current(). 2019-03-10 20:38:28 +00:00
5071baa2aa Print window id in hex; while here, remove unnecessary newline. 2019-03-08 20:33:30 +00:00
4470a247c8 Similar to keypress event, fetch the screen from the event root window in the
buttonpress handler; bail if we don't manage the screen. Allows us to find the
current client based on the screen/event root.
2019-03-08 17:40:43 +00:00
eab4b7e4b5 extend verbose logging for key/button events 2019-03-08 15:04:39 +00:00
bf43b62414 [keypress event] turns out we've been checking the wrong window for a matching
client thus always falling back to client_current(); while the current client
is problaby right in most cases, use event's subwindow (not window) to find the
client. Bail early if this event came to us from a screen we don't manage.
This is result of us grabing all keybindings off the root window instead of
selectively.
2019-03-08 14:48:02 +00:00
2a3c2b5231 add parans for readibility 2019-03-08 13:17:26 +00:00
7c45b87622 Teach client_current() to use a screen to find the current client instead of
iterating over all (fallback if no screen provided for now). Initially convert
trivial uses of client_current().
2019-03-07 14:28:17 +00:00
01be5b4e4a check cc->gc directly 2019-03-07 13:24:44 +00:00
823566a653 zip extra lines 2019-03-07 13:24:10 +00:00
aa79351d2e gc clientq inside groups, instead use the better maintained one per-screen 2019-03-07 13:14:41 +00:00
b26202724a shuffle deck chairs: rename group actions to match intent for clarity 2019-03-07 12:54:21 +00:00
9efa6c8c85 same thing as screen_find() 2019-03-06 13:32:19 +00:00
0bda8f7606 Separate out the menu window from the client resize/move geom window; in each
case, create and destroy on-demand. Isolate more menu specific code.
2019-03-04 19:28:17 +00:00