bspwm — A tiling window manager based on binary space partitioning

pa1adin · 2014-07-28 19:14:08

The easiest and dirty solution what i've found, start synapse from bspwmrc:

synapse && bspc config -w focused border_width 0

Works.

o_caino · 2014-07-28 19:17:31

pa1adin wrote:

start synapse from bspwmrc

Or the same command in sxhkdrc.

pa1adin · 2014-07-28 19:19:18

Much better!.

vorenon · 2014-07-30 21:25:46

Since upgrading to xorg-server 1.16, it's hard to restart bspwm a second time because it segfaults nearly every time.

After deleting the socket file in /tmp, it works fine, so it looks like a permission error. Any ideas how to fix this?

Stebalien · 2014-07-31 04:28:38

vorenon wrote:

Since upgrading to xorg-server 1.16, it's hard to restart bspwm a second time because it segfaults nearly every time.
After deleting the socket file in /tmp, it works fine, so it looks like a permission error. Any ideas how to fix this?

Report an issue on github and provide a gdb stack trace.

vorenon · 2014-07-31 12:49:34

Stebalien wrote:

vorenon wrote:
Since upgrading to xorg-server 1.16, it's hard to restart bspwm a second time because it segfaults nearly every time.
After deleting the socket file in /tmp, it works fine, so it looks like a permission error. Any ideas how to fix this?
Report an issue on github and provide a gdb stack trace.

Done. I don't know how gdb works so I hope a strace log is enough.

smlb · 2014-08-01 09:35:24

bloom: actually my blog is down (http://smlb.pw/bspwm-how-to/), I'll write the same How-To on my github and I'll link here. Sorry!

EDIT: here you are the full guide https://github.com/smlb/dotfiles/wiki/Bspwm

Last edited by smlb (2014-08-01 10:26:08)

bloom · 2014-08-01 12:14:36

smlb wrote:

here you are the full guide https://github.com/smlb/dotfiles/wiki/Bspwm

Thanks.

Stalafin · 2014-08-01 15:02:57

When I kill xinit or bspwm, the panel (bar-aint-recursive) copied from the baskerville's / bloom's example does not get killed with it. It just stays alive and keeps running in the background, even without X (while maxing out my CPU). So what I usually do is to kill the panel by hand afterwards.

What I find surprising is that the panel has a different PGID from bspwm, even though it is started from within bspwm:

 PGID   PID COMMAND         CMD
  793   793 startx          /bin/sh /usr/bin/startx
  793   927  \_ xinit        \_ xinit /home/$USER/.xinitrc -- /etc/X11/xinit/xserverrc :0 -auth /tmp/serverauth.GubNWU53XI
  928   928      \_ Xorg.bi      \_ /usr/bin/Xorg.bin -nolisten tcp :0 -auth /tmp/serverauth.GubNWU53XI vt1
  935   935      \_ wm           \_ /bin/sh /home/$USER/.local/bin/wm
  935   941          \_ xsc          \_ /usr/bin/xscreensaver -no-splash
  935   942          \_ lig          \_ /bin/bash /home/$USER/.local/bin/lightsOn.sh
  935 23300          |   \_          |   \_ sleep 290
  935   947          \_ sxh          \_ /usr/bin/sxhkd
  935   951          \_ bsp          \_ bspwm
  955   983 panel           /bin/sh /home/$USER/.local/bin/panel
  955   986  \_ bspc         \_ bspc control --subscribe
  955   987  \_ xtitle       \_ xtitle -sf T%s
  955   988  \_ clock        \_ clock -sf C%b %d | %H:%M
  955   989  \_ battery      \_ battery -sf Bs%s
  955   990  \_ battery      \_ battery -sf Bl%i
  955   991  \_ cat          \_ cat /tmp/panel-fifo
  955   992  \_ panel_bar    \_ /bin/sh /home/$USER/.local/bin/panel_bar
  955   993  \_ bar-aint-re  \_ bar-aint-recursive -f -*-terminus-medium-r-normal-*-12-*-*-*-c-*-*-* -B #FF34322E -F #FFA3A6AB -g x24
  955   994  \_ panel        \_ /bin/sh /home/$USER/.local/bin/panel

I guess this line in the configuration should make sure that everything gets killed cleanly, but I guess if it doesn't fire kill 0, it means that panel never receives a SIG{KILL,TERM,QUIT,INT} in the first place:

trap 'trap - TERM; kill 0' INT TERM QUIT EXIT

Any idea what could be happening?

smlb · 2014-08-01 15:26:56

Stalafin, add this to your sxhkdrc

# This
bspc quit && pkill panel 
# Instead of this
bspc quit

Last edited by smlb (2014-08-01 15:28:19)

Stalafin · 2014-08-01 15:47:25

smlb wrote:

Stalafin, add this to your sxhkdrc
# This
bspc quit && pkill panel 
# Instead of this
bspc quit

That is certainly the proper way of shutting down bspwm.

What just surprises me is that if I kill the xinit process itself (pkill -x xinit), the panel keeps on happily running. I wonder why that is.

Last edited by Stalafin (2014-08-01 15:47:41)

Dragon707 · 2014-08-01 15:59:00

Stalafin wrote:

smlb wrote:
Stalafin, add this to your sxhkdrc
# This
bspc quit && pkill panel 
# Instead of this
bspc quit
That is certainly the proper way of shutting down bspwm.
What just surprises me is that if I kill the xinit process itself (pkill -x xinit), the panel keeps on happily running. I wonder why that is.

I guess thats it because its only running the script, the bar or dzen just stops if im correct.

Stalafin · 2014-08-01 16:41:28

Dragon707 wrote:

Stalafin wrote:
smlb wrote:
Stalafin, add this to your sxhkdrc
# This
bspc quit && pkill panel 
# Instead of this
bspc quit
That is certainly the proper way of shutting down bspwm.
What just surprises me is that if I kill the xinit process itself (pkill -x xinit), the panel keeps on happily running. I wonder why that is.
I guess thats it because its only running the script, the bar or dzen just stops if im correct.

You are not correct. The panel stays alive, as does bar-aint-recursive (which is maxing out my CPU).

jan · 2014-08-07 11:44:45

Is there a way to focus window on current monitor? something like

bspc window -f right.current_monitor

Last edited by jan (2014-08-07 11:45:30)

Stebalien · 2014-08-08 04:38:40

jan wrote:

Is there a way to focus window on current monitor? something like
bspc window -f right.current_monitor

bspc window -f right.local

?

jan · 2014-08-08 06:59:05

Stebalien wrote:

jan wrote:
Is there a way to focus window on current monitor? something like
bspc window -f right.current_monitor
bspc window -f right.local
?

Are you sure? In my test, .local selects window in current 'desktop', not 'monitor'

o_caino · 2014-08-08 09:12:13

btw, I noticed that the command

bspc window -f {next,prev}

does the same as

bspc window -f {next,prev}.local

both select only windows on the current desktop.

Stebalien · 2014-08-09 05:24:06

jan wrote:

Stebalien wrote:
jan wrote:
Is there a way to focus window on current monitor? something like
bspc window -f right.local
?
Are you sure? In my test, .local selects window in current 'desktop', not 'monitor'

Please define "windows in current monitor" (I define it as windows visible on the current monitor). Do you want to switch between all windows on all desktops in the current monitor? This doesn't make sense because desktops aren't spatially oriented with respect to each other. That is, there is no concept "the desktop to the right of this one", there is only next and previous.

o_caino wrote:

btw, I noticed that the command
bspc window -f {next,prev}
does the same as
bspc window -f {next,prev}.local
both select only windows on the current desktop.

The first one will jump between monitors, the second will stay within the confines of the current monitor.

Last edited by Stebalien (2014-08-09 05:26:19)

o_caino · 2014-08-09 08:40:53

Stebalien wrote:

The first one will jump between monitors, the second will stay within the confines of the current monitor.

Not in my case. Both behave the same way: focus only windows on the current desktop.

Last edited by o_caino (2014-08-09 08:41:40)

jan · 2014-08-09 12:58:32

Thanks @Stebalien and @o_caino! Your opposite testimony on '.local' behavior made me think it could be a version problem. So I did 'sudo yaourt -S bspwm-git' to upgrade, and '.local' works as @Stebalien described now :-)

angelic_sedition · 2014-08-09 18:14:09

I've been messing around with herbstluftwm today. The hooks have given me some cool ideas for using

bspc control --subscribe

I also noticed that I had background window opacity set in my compton.conf. Compton doesn't seem to be able to detect inactive windows with bspwm though. The only thing I could find about this was this reddit post (with no solution). Does anyone here know how to get compton to recognize unfocused windows with bspwm?

Last edited by angelic_sedition (2014-08-09 18:14:27)

cju · 2014-08-09 21:28:28

angelic_sedition wrote:

Compton doesn't seem to be able to detect inactive windows with bspwm though.

Nope, that should work indeed. Please post your compton config for further help.

o_caino · 2014-08-10 11:12:39

jan wrote:

Your opposite testimony on '.local' behavior made me think it could be a version problem.

I check everyday for updates, so I'm always running the most recent (-git) version of both bspwm and sxhkd...

jan wrote:

In my test, .local selects window in current 'desktop', not 'monitor'

Stebalien wrote:

The first one will jump between monitors, the second will stay within the confines of the current monitor.

Note that there are two different .local modifiers

man bspwm wrote:

Window
Select a window.
WINDOW_SEL := <window_id>
| (DIR|CYCLE_DIR|biggest|last|focused|older|newer)[.float
ing|.tiled][.like|.unlike][.manual|.automatic][.urgent][.local]
[...]
local
Only consider windows of the current desktop.
[...]
Desktop
Select a desktop.
DESKTOP_SEL := <desktop_name>
| [MONITOR_SEL:]^<n>
| (CYCLE_DIR|last|[MONITOR_SEL:]focused|older|newer)[.occupied|.free][.urgent][.local]
[...]
local
Only consider desktops of the current monitor.

o_caino wrote:

bspc window -f {next,prev}.local

I'm talking about the former.
If there's a modifier that restrics matches to the current desktop I'd expect the default behaviour, without modifiers, to match windows regardless of the desktop they're in. And, as I said on a previous post, this is not the case (at least for me).

Last edited by o_caino (2014-08-10 11:19:59)

angelic_sedition · 2014-08-11 18:24:43

cju wrote:

angelic_sedition wrote:
Compton doesn't seem to be able to detect inactive windows with bspwm though.
Nope, that should work indeed. Please post your compton config for further help.

I should have checked first all my settings; the problem was in my config:

mark-wmwin-focused = true;

This seems to have no effect in herbstluftwm and a few others I tried, but with bspwm all windows get marked as focused I guess, so it has to be set to false.

cju · 2014-08-12 10:19:18

angelic_sedition wrote:

I should have checked first all my settings; the problem was in my config:
mark-wmwin-focused = true;
This seems to have no effect in herbstluftwm and a few others I tried, but with bspwm all windows get marked as focused I guess, so it has to be set to false.

You don't have to set it to false, just don't set it at all since it's false by default. It's strange though cause I don't run into that issue having wmwin enabled… Which versions of bspwm and compton do you use?

Arch Linux

#1351 2014-07-28 19:14:08

Re: bspwm — A tiling window manager based on binary space partitioning

#1352 2014-07-28 19:17:31

Re: bspwm — A tiling window manager based on binary space partitioning

#1353 2014-07-28 19:19:18

Re: bspwm — A tiling window manager based on binary space partitioning

#1354 2014-07-30 21:25:46

Re: bspwm — A tiling window manager based on binary space partitioning

#1355 2014-07-31 04:28:38

Re: bspwm — A tiling window manager based on binary space partitioning

#1356 2014-07-31 12:49:34

Re: bspwm — A tiling window manager based on binary space partitioning

#1357 2014-08-01 09:35:24

Re: bspwm — A tiling window manager based on binary space partitioning

#1358 2014-08-01 12:14:36

Re: bspwm — A tiling window manager based on binary space partitioning

#1359 2014-08-01 15:02:57

Re: bspwm — A tiling window manager based on binary space partitioning

#1360 2014-08-01 15:26:56

Re: bspwm — A tiling window manager based on binary space partitioning

#1361 2014-08-01 15:47:25

Re: bspwm — A tiling window manager based on binary space partitioning

#1362 2014-08-01 15:59:00

Re: bspwm — A tiling window manager based on binary space partitioning

#1363 2014-08-01 16:41:28

Re: bspwm — A tiling window manager based on binary space partitioning

#1364 2014-08-07 11:44:45

Re: bspwm — A tiling window manager based on binary space partitioning

#1365 2014-08-08 04:38:40

Re: bspwm — A tiling window manager based on binary space partitioning

#1366 2014-08-08 06:59:05

Re: bspwm — A tiling window manager based on binary space partitioning

#1367 2014-08-08 09:12:13

Re: bspwm — A tiling window manager based on binary space partitioning

#1368 2014-08-09 05:24:06

Re: bspwm — A tiling window manager based on binary space partitioning

#1369 2014-08-09 08:40:53

Re: bspwm — A tiling window manager based on binary space partitioning

#1370 2014-08-09 12:58:32

Re: bspwm — A tiling window manager based on binary space partitioning

#1371 2014-08-09 18:14:09

Re: bspwm — A tiling window manager based on binary space partitioning

#1372 2014-08-09 21:28:28

Re: bspwm — A tiling window manager based on binary space partitioning

#1373 2014-08-10 11:12:39

Re: bspwm — A tiling window manager based on binary space partitioning

#1374 2014-08-11 18:24:43

Re: bspwm — A tiling window manager based on binary space partitioning

#1375 2014-08-12 10:19:18

Re: bspwm — A tiling window manager based on binary space partitioning

Board footer