bspwm — A tiling window manager based on binary space partitioning

egroeg · 2014-01-23 01:08:31

ict wrote:

Yes, # starts a comment in bash.

doh! thanks! ...man, i feel silly now.

senft · 2014-01-28 21:25:25

Hey,
not sure if this is a bug. Some config commands, when called without an argument, return the currently set value. Some others dont. For example running:

# bspc config split_ratio

prints "0.520000". Running

# bspc config window_gap

prints nothing. It would be nice to have a way of getting the currently window_gap or paddings.
I tried to dig in the code, but I suck at C
Best, senft

Last edited by senft (2014-01-28 21:26:25)

aoba · 2014-01-28 22:01:59

Looks like it is probably a bug in cmd_config in messages.c. Here is a possibly corrected version:

bool cmd_config(char **args, int num, char *rsp)
{
	if (num < 1)
		return false;
	coordinates_t ref = {mon, mon->desk, mon->desk->focus};
	coordinates_t trg = {NULL, NULL, NULL};
	if ((*args)[0] == OPT_CHR) {
		if (streq("-d", *args) || streq("--desktop", *args)) {
			num--, args++;
			if (num < 1)
				return false;
			if (!desktop_from_desc(*args, &ref, &trg))
				return false;
		} else if (streq("-m", *args) || streq("--monitor", *args)) {
			num--, args++;
			if (num < 1)
				return false;
			if (!monitor_from_desc(*args, &ref, &trg))
				return false;
		} else {
			return false;
		}
		num--, args++;
	} else {		/* changes here */
		trg = ref;
	}			

	if (num == 2)
		return set_setting(trg, *args, *(args + 1));
	else if (num == 1)
		return get_setting(trg, *args, rsp);
	else
		return false;
}

(I say probably because I can see what's wrong - in the original, trg is full of NULLs if --desktop or --monitor is not used - but I don't really know if the trg = ref; assigment is "good" C code.)

Edit: for now you should be able to use, for example, bspc config --desktop focused window_gap .

Last edited by aoba (2014-01-28 22:04:30)

earsplit · 2014-01-29 00:24:46

senft wrote:

Hey,
not sure if this is a bug. Some config commands, when called without an argument, return the currently set value. Some others dont. For example running:
# bspc config split_ratio
prints "0.520000". Running
# bspc config window_gap
prints nothing. It would be nice to have a way of getting the currently window_gap or paddings.
I tried to dig in the code, but I suck at C
Best, senft

Window gap and padding are per-desktop settings. If you change it on one desktop, bspc config window_gap would be undeterminate.

As pointed out above, you can use

 bspc config -d focused window_gap

to get this value, as in

# Change window gap
alt + {minus,equal}
        bspc config -d focused window_gap $((`bspc config -d focused window_gap` {+,-} 2 ))

aoba · 2014-01-29 04:44:50

earsplit wrote:

Window gap and padding are per-desktop settings. If you change it on one desktop, bspc config window_gap would be undeterminate.

Ah. I assumed that the currently-focused desktop/monitor was supposed to be the implicit target if one was not specified. Your explanation makes more sense.

senft · 2014-01-29 08:44:37

Ah, thanks, both of you guys!
Maybe an error message (missing target: specify a desktop or something like that) would be useful?
Somewhat Offtopic: How would I tinker around with the code? When I compile bspwm, do I have to restart X and start the frehsly compiled version of bspwm? Other WMs often seem to have some kind of mechanism to replace the currently running wm (--replace).

enr1x · 2014-01-29 12:26:01

senft wrote:

Ah, thanks, both of you guys!
When I compile bspwm, do I have to restart X and start the frehsly compiled version of bspwm? Other WMs often seem to have some kind of mechanism to replace the currently running wm (--replace).

You don't have to restart it. You can just kill the currently running wm instance and then fire up a new instance of the WM.

senft · 2014-01-29 12:52:33

I tried that with

killall bspwm && ./bspwm

But then the complete X session seems to be restarted (and all apps are killed, which is tedious while developing. I would have to restart all editors and stuff...) Is there something wrong with my .xinitrc? I start bspwm like this:

sxhkd -f 100 &
exec bspwm

Last edited by senft (2014-01-29 12:53:44)

bloom · 2014-01-29 13:00:08

senft wrote:

When I compile bspwm, do I have to restart X and start the frehsly compiled version of bspwm?

Have a look at this.

shroomin · 2014-01-31 19:00:06

Hi!
Sorry if someone had this issue before, couldn't find anything.
If I make a window fullscreen, suddenly all the floating windows on all desktops (the ones that were before and ones that are opened after) start to appear behind the tiled ones and it persists until all windows are closed/moved from said desktop (the one that had a fullscreen window on it).
Tl;dr: maximize a window on a desktop X -> all the floating windows everywhere switch behind tiled ones until desktop X emptied.

bloom · 2014-01-31 20:20:25

shroomin wrote:

If I make a window fullscreen, suddenly all the floating windows on all desktops (the ones that were before and ones that are opened after) start to appear behind the tiled ones [...].

Thanks for reporting!

It should be fixed by f5054b5.

o_caino · 2014-02-05 01:57:59

Hi,

Since moving from rulc and ruld to built-in rules I'm experiencing a strange issue with Transmission-gtk. I have the following rule for Transmission-gtk

bspc rule -a Transmission-gtk desktop=^9

When I run Transmission-gtk it follows the window to desktop 9 as if follow=on. I tried appending follow=off, just in case, but it doesn't work.
Is this a bug or is my config wrong?

The class name reported by the command bspc query -T and xprop is the same as the rule.
All my rules for other applications work.

Thanks!

agahnim · 2014-02-05 22:26:42

bspc rule -a Steam floating=on desktop=^3
bspc rule -a Gimp floating=on
bspc rule -a Pidgin desktop=^2

I have an issue with rules as well. I believe these to be correct, as a friend of mine has the steam rule verbatum and it works for him and not for me. I have also checked the class names with bspc query -T and have come up with these names being the correct ones. Hell the gimp rule is straight from the example config. I have the optional dependancies installed, even though built in rules are not supposed to require anything additional. I am also using the git version by the way. Any suggestions?

cju · 2014-02-06 12:39:41

I have a general question: Is it possible in bspwm to move/resize floating windows manually with the keyboard, not just with the mouse? For example like one can do that with i3…

tonk · 2014-02-06 14:08:56

cju wrote:

I have a general question: Is it possible in bspwm to move/resize floating windows manually with the keyboard, not just with the mouse? For example like one can do that with i3…

#p1364629
but haven't tried it by myself.

o_caino wrote:

When I run Transmission-gtk it follows the window to desktop 9 as if follow=on. I tried appending follow=off, just in case, but it doesn't work.

you could try

bspc config ignore_ewmh_focus true

cju · 2014-02-06 14:42:17

tonk wrote:

cju wrote:
I have a general question: Is it possible in bspwm to move/resize floating windows manually with the keyboard, not just with the mouse? For example like one can do that with i3…
#p1364629
but haven't tried it by myself.

Thank you, that was exactly what I was looking for. Works like a charm btw.

bloom · 2014-02-06 20:09:32

tonk wrote:

o_caino wrote:
When I run Transmission-gtk it follows the window to desktop 9 as if follow=on. I tried appending follow=off, just in case, but it doesn't work.
you could try
bspc config ignore_ewmh_focus true

As of fd2a62a, it will, indeed, work as intended (i.e. pagers and switchers will continue to work).

o_caino · 2014-02-06 22:26:43

Updated to latest version and added

bspc config ignore_ewmh_focus true

to config.
Now it works.

Thanks!

Šaran · 2014-02-07 09:35:31

Hi guys,

I have just started playing with bspwm and I have a question regarding multihead setup.

I have two monitors, LVDS-0 and VGA-0, configured by

bspc monitor LVDS-0 -d 1 2 3 4 5
bspc monitor VGA-0 -d 6 7 8 9 10

Problem is LVDS-0 desktops get assigned keybinds 6-10, instead of 1-5 and desktops in panel look like this:

6 7 8 9 10 1 2 3 4 5

$ xrandr; bspc query -T

Screen 0: minimum 8 x 8, current 3200 x 1080, maximum 16384 x 16384
VGA-0 connected 1280x1024+1920+0 (normal left inverted right x axis y axis) 338mm x 270mm
   1280x1024      60.0*+   75.0  
   1024x768       75.0     60.0  
   800x600        75.0     60.3  
   640x480        75.0     59.9  
LVDS-0 connected primary 1920x1080+0+0 (normal left inverted right x axis y axis) 382mm x 215mm
   1920x1080      60.0*+
HDMI-0 disconnected (normal left inverted right x axis y axis)
HDMI-1 disconnected (normal left inverted right x axis y axis)
VGA-0 1280x1024+1920+0 24,0,0,0
	6 2 5 0,0,0,0 T - *
	7 2 5 0,0,0,0 T -
	8 2 5 0,0,0,0 T -
	9 2 5 0,0,0,0 T -
	10 2 5 0,0,0,0 T -
LVDS-0 1920x1080+0+0 24,0,0,0 *
	1 2 5 0,0,0,0 T - *
		V m 0.500000
			m URxvt urxvt 0x1600016 2 804x436+0+0 L -------- *
			m Firefox Navigator 0x260008B 2 949x1042+962+29 R --------
	2 2 5 0,0,0,0 T -
	3 2 5 0,0,0,0 T -
	4 2 5 0,0,0,0 T -
	5 2 5 0,0,0,0 T -

How do I make bspwm treat my main monitor as primary?

Thanks!

Last edited by Šaran (2014-02-07 09:37:43)

bloom · 2014-02-07 11:04:25

Šaran wrote:

Problem is LVDS-0 desktops get assigned keybinds 6-10, instead of 1-5 and desktops in panel look like this:
6 7 8 9 10 1 2 3 4 5

It should be fixed by 764b428.

Šaran · 2014-02-08 11:03:07

Thank you! It's working as expected now.

cju · 2014-02-08 16:23:20

Perhaps this is stupid question, but is there a way to get all floating windows centered (instead of positioned at 0,0) by default, no matter if the window was originally floating by a rule or just toogled into floating?

bloom · 2014-02-08 18:05:27

cju wrote:

Is there a way to get all floating windows centered (instead of positioned at 0,0) by default, no matter if the window was originally floating by a rule or just toogled into floating?

It should be happening as of 167bd82.

cju · 2014-02-08 18:42:23

bloom wrote:

cju wrote:
Is there a way to get all floating windows centered (instead of positioned at 0,0) by default, no matter if the window was originally floating by a rule or just toogled into floating?
It should be happening as of 167bd82.

Awesome! Thanks a lot.

cju · 2014-02-08 22:35:08

cju wrote:

bloom wrote:
cju wrote:
Is there a way to get all floating windows centered (instead of positioned at 0,0) by default, no matter if the window was originally floating by a rule or just toogled into floating?
It should be happening as of 167bd82.
Awesome! Thanks a lot.

I'm sorry that I have to bring this up again, but it's a bit strange: now, as said before, floating windows are centered automatically, but they aren't centered correctly. I.e. there is quite more screenspace on the left and on the top than on the right and on the bottom.

You can observe this wrong positioning better if you open a normal window (which is "centered" correctly) and increase window_gap until it has nearly the dimensions of a floating window. Then open a floating window and since it overlaps the previous one, you can see the off-center position of the floating one. It's also pretty visible if you send big windows like firefox to floating, beause they aren't decreased in size that much and so the different spaces to the borders of the monitor become more visible… Or just take a ruler.

Arch Linux

#1051 2014-01-23 01:08:31

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

#1052 2014-01-28 21:25:25

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

#1053 2014-01-28 22:01:59

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

#1054 2014-01-29 00:24:46

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

#1055 2014-01-29 04:44:50

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

#1056 2014-01-29 08:44:37

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

#1057 2014-01-29 12:26:01

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

#1058 2014-01-29 12:52:33

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

#1059 2014-01-29 13:00:08

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

#1060 2014-01-31 19:00:06

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

#1061 2014-01-31 20:20:25

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

#1062 2014-02-05 01:57:59

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

#1063 2014-02-05 22:26:42

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

#1064 2014-02-06 12:39:41

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

#1065 2014-02-06 14:08:56

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

#1066 2014-02-06 14:42:17

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

#1067 2014-02-06 20:09:32

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

#1068 2014-02-06 22:26:43

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

#1069 2014-02-07 09:35:31

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

#1070 2014-02-07 11:04:25

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

#1071 2014-02-08 11:03:07

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

#1072 2014-02-08 16:23:20

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

#1073 2014-02-08 18:05:27

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

#1074 2014-02-08 18:42:23

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

#1075 2014-02-08 22:35:08

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

Board footer