Re: [dwm] Mouse over tile ...
Tako rzecze Anselm R. Garbe (w e-mailu datowanym 2007-09-16, 14:34): So there arises a question. Traditionally in dwm Mod-Button2 was introduced for zooming clients in tiled mode. But now I more and more come to the conclusing, that this should be togglefloating() instead. Is there anyone really using Mod-Button2 for zoom'ing? Honestly -- first time I've heard about it. Go ahead. [a] -- . Antoni Grzymala - antoni (at) chopin.edu.pl ---. | OpenPGP KeyID EB315583 available now from a keyserver near you | | Fingerprint A819 6D2E D5EB D9E0 D2D9 7AF6 2FAF 4A11 EB31 5583 | `' pgphAONpfRWns.pgp Description: PGP signature
Re: [dwm] Mouse over tile ...
I think it's really useful. However I also implemented Mod-Button2 for toggling such floating clients back to tiled. Sorry, part of a previous message was nonsense. -- Damjan Vrenčur ~ http://lmmri.fri.uni-lj.si/damjan/ ~ GPG key: C6A3146F
Re: [dwm] Mouse over tile ...
On (16/09/07 15:13), Damjan Vrencur wrote: But since Mod-Button1 does the same provided you don't move mouse, isn't it better to use Mod-Button2 for something else? I checked the latest tip and it seems to work nice except that now I got some unused space between clients. Screenshot: http://lmmri.fri.uni-lj.si/damjan/dwmScreenshot.png Peace, change: #define RESIZEHINTS True in config.h to: #define RESIZEHINTS False -- Premysl Anydot Hruby http://na.srck.net
Re: [dwm] Mouse over tile ...
I have always find the mod-click2 feature quite useful and handy, I have been using it since the beggining. Please do not remove it. Mod-Click1 to make windows floating looks cleaner to me. On Sun, 16 Sep 2007 15:13:17 +0200 Damjan Vrencur [EMAIL PROTECTED] wrote: On Sunday 16 of September 2007 14:48:22 Antoni Grzymala wrote: Honestly -- first time I've heard about it. Me too. But since Mod-Button1 does the same provided you don't move mouse, isn't it better to use Mod-Button2 for something else? I checked the latest tip and it seems to work nice except that now I got some unused space between clients. Screenshot: http://lmmri.fri.uni-lj.si/damjan/dwmScreenshot.png Peace, -- Damjan Vrenčur ~ http://lmmri.fri.uni-lj.si/damjan/ ~ GPG key: C6A3146F --pancake
[dwm] Unused space between clients
On Sun, Sep 16, 2007 at 03:13:17PM +0200, Damjan Vrencur wrote: I checked the latest tip and it seems to work nice except that now I got some unused space between clients. Screenshot: http://lmmri.fri.uni-lj.si/damjan/dwmScreenshot.png That space is much smaller here. And this is actually a big improvment. Previously, I had some weird corruptions at the place of this unused space. This only happened with urxvt though, not xterm. Not sure why.. But I'm happier with urxvt anyway. What I'm wondering is if it wouldn't be possible to only have unused space (when necessary) at the bottom of the column, instead of between each clients, while still fixing that same corruption bug of course.
Re: [dwm] Updated bottom stack patch
On Sun, Sep 16, 2007 at 01:18:24PM +0200, Anselm R. Garbe wrote: Hi James, cool, I want to give an example how your patch should be reorganized that it works with less of a hassle beginning with dwm-4.5: [snip] Thanks for the great example, that will especially make things easier while the code is still being changed. I think before 4.4 was released I probably rewrote my patch at least a dozen times to work with the changing source tree. This of course will eliminate that need all together. -- James Turner BSD Group Consulting http://www.bsdgroup.org
Re: [dwm] Proper way of monitoring battery level in DWM
MAX_BATTERY_CAPACITY=84240
while true
do
echo `date '+%Y%m%d %H:%M'` `uptime | sed 's/.*:
\(.\...\),.*/\1/'` $(( (`cat /proc/acpi/battery/BAT0/state | grep
remaining capacity | sed 's/.*: \{5,\}\(.*\) mWh/\1/'` * 100) /
$MAX_BATTERY_CAPACITY))%
sleep 20
done | /home/albert/Programming/suckless/bin/dwm
Beware of email's silly line breaks.
--
Albert Cardona
http://www.mcdb.ucla.edu/Research/Hartenstein/acardona
[dwm] Fwd: Auto-discard notification
The attached message has been automatically discarded. arg: On Sun, Sep 16, 2007 at 03:51:19PM +0200, pancake wrote: I have always find the mod-click2 feature quite useful and handy, I have been using it since the beggining. Please do not remove it. Mod-Click1 to make windows floating looks cleaner to me. Well there must be a simple way to put a temporarily moved _tiled_ window back to tiled state again. So Mod-Button2 seems natural for this. xmonad uses mod-t for re-tiling these guys, but it seems like a nice idea to use a mouse button. -- Don - End forwarded message - -- Anselm R. Garbe http://www.suckless.org/ GPG key: 0D73F361
Re: [dwm] I merged the whole source to one file and wanna let you know the implications
That works, and I agree that's simpler. On Sun, 2007-09-16 at 13:06 +0200, Anselm R. Garbe wrote: I think one can argue a lot about how to organize source code the right way. Especially if your codebase is huge. Anyway, I realized that I reorganized the dwm code base a couple of times, and this has been due to one fact: dwm's code base has been organized following two different paradigms - separating code into logically equal parts on one side (client.c, event.c, screen.c, layout.c, tag.c, ..) and functionally equal parts (draw.c, main.c, util.c)i on the other side... Due to this separation often the need arised to enforce consistency into the one or the other direction. I believe that also other projects who thought carefully about these questions noticed the same. However, I browsed through the old Unix tools several times, and they often cosist of only a single source file. That's because they want to meet the goal do only ONE thing, and do it well. So organizing dwm in a single dwm.c (see hg tip) will enforce to reach the ideal much more in my conlusion. There are also some positive benefits in this decision. The overhead to differentiate between static and extern variables/functions is gone now, you can call all functions and use all global variables within dwm.c. config.h can refer to any global variable or function now. This includes extending dwm with different layouts - you only need to write a .c file now (no need to write a .h file), include that .c file in config.h right before the Layout-array initialization and it will simply work. You can declare all functions static in external layout implementations, simply because they will be part of dwm.c anyways after CPP did its job. I will do the same with dmenu as well. Regards, Lobzang ___ http://lobzang.free.fr
Re: [dwm] Proper way of monitoring battery level in DWM
Thank you for all the responses. Actually right now, I am using apm. I am not sure if acpi is supported on my notebook. This is an old PowerBook G4 500MHz. I will try modifying the scripts to use apm instead of acpi. Amit
Re: [dwm] Proper way of monitoring battery level in DWM
On 9/16/07, Amit [EMAIL PROTECTED] wrote: Actually right now, I am using apm. I am not sure if acpi is supported on my notebook. This is an old PowerBook G4 500MHz. I will try modifying the scripts to use apm instead of acpi. If it's a PowerBook G4, then it's using PMU, not ACPI nor APM. Under GNU/Linux, you can use the command `pbbcmd` to query the battery status:: $ pbbcmd query powersource 0 # -- outputs 0 if running on battery, and 1 if running on AC power. :: $ pbbcmd query timeremaining 8760 # -- remaining running time in seconds. Or you can directly parse the /proc/pmu/battery_0 file:: $ cat /proc/pmu/battery_0 flags : 0011 charge : 3251 max_charge : 3251 current: -1334 voltage: 12613 time rem. : 8773 For instance you could define a `battstat` command:: #! /bin/sh if [ -r /proc/pmu/battery_0 ]; then sed -e 's/ *: */ /' /proc/pmu/battery_0 | ( \ while read key value; do case $key in charge) charge=$value;; max_charge) max_charge=$value;; current) if [ $value -eq 0 ]; then on_ac_power=yes; fi;; esac done if [ -z $on_ac_power ]; then echo batt:$(expr $charge \* 100 / $max_charge )% else echo (AC) fi ) fi And then invoke dwm like:: (while sleep 10; do battstat; done) | dwm Riccardo
Re: [dwm] Proper way of monitoring battery level in DWM
On Sep 16, 2007, at 4:34 PM, Amit wrote: Thank you for all the responses. Actually right now, I am using apm. I am not sure if acpi is supported on my notebook. This is an old PowerBook G4 500MHz. I will try modifying the scripts to use apm instead of acpi. Amit I just use something like this echo `/usr/sbin/apm -l`%. If all you want is the remaining percentage it does the trick nicely. -- James Turner BSD Group Consulting http://www.bsdgroup.org
[dwm] [patch] Multihead Layout
I often connect my laptop to an external monitor and use it to extend my
desktop. This patch adds a new layout 'multihead' which uses Xrandr to detect
the configuration of monitors, and adjusts window sizes accordingly.
Here is my X screen layout:
--
| | |
| | |
--+ |
The left screen is the internal (widescreen) monitor on my laptop with 1280x800
resolution. The right one is my external monitor with 1280x1024 resolution. The
standard dwm tile layout has the problem that it uses the dead space underneath
the left screen, and I am unable to see anything there. This patch detects how
many monitors I have connected via a call to xrandr, and tiles windows so that
they fit fullscreen in each window. As soon as you have more windows than
monitors, one monitor is used for stacking and the others are used for
fullscreen.
There is a lot of missing functionality:
* Add the ability to move a window to different screens. Right now you are
stuck with where this layout places them.
* Don't re-detect monitor layout each time we tile windows. Instead use the
Xrandr event mechanisms.
* Disable this layout automatically when there is only one monitor connected.
* Decide on the order of monitors based on x/y positions instead of whatever
order we get them from Xrandr.
The patch is against the latest hg tip (with all sources merged into one file).
I'm pretty new to this so I welcome even the most basic feedback.
You will need the Xrandr extension and dev headers to compile this patch
(apt-get install libxrandr-dev in Ubuntu).
Cheers,
Tim
diff -r d036b2f17567 config.h
--- a/config.h Sun Sep 16 13:42:37 2007 +0200
+++ b/config.h Sun Sep 16 11:27:16 2007 -0700
@@ -21,9 +21,12 @@ static Rule rules[] = {
{ Acroread, NULL, True },
};
+#include multihead.c
+
/* layout(s) */
static Layout layouts[] = {
/* symbol function */
+ { UU=,multihead }, /* first entry is default */
{ []=,tile }, /* first entry is default */
{ ,floating },
};
diff -r d036b2f17567 config.mk
--- a/config.mk Sun Sep 16 13:42:37 2007 +0200
+++ b/config.mk Sun Sep 16 11:27:16 2007 -0700
@@ -12,7 +12,7 @@ X11LIB = /usr/X11R6/lib
# includes and libs
INCS = -I. -I/usr/include -I${X11INC}
-LIBS = -L/usr/lib -lc -L${X11LIB} -lX11
+LIBS = -L/usr/lib -lc -L${X11LIB} -lX11 -lXrandr
# flags
CFLAGS = -Os ${INCS} -DVERSION=\${VERSION}\
diff -r d036b2f17567 multihead.c
--- /dev/null Thu Jan 01 00:00:00 1970 +
+++ b/multihead.c Sun Sep 16 11:27:16 2007 -0700
@@ -0,0 +1,41 @@
+/* See LICENSE file for copyright and license details. */
+#include stdio.h
+#include X11/extensions/Xrandr.h
+
+void
+multihead(void) {
+ unsigned int i, n, nw, nh, nx, ny;
+ Client *c;
+ XRRScreenResources *res;
+ XRRCrtcInfo *crtc_info, *last_crtc_info;
+
+ for(n = 0, c = nexttiled(clients); c; c = nexttiled(c-next))
+ n++;
+
+ res = XRRGetScreenResources(dpy, root);
+ last_crtc_info = XRRGetCrtcInfo(dpy, res, res-crtcs[res-ncrtc-1]);
+
+ for(i = 0, c = nexttiled(clients); c; c = nexttiled(c-next), i++) {
+ c-ismax = False;
+ if(i res-ncrtc - 1) { /* master */
+ crtc_info = XRRGetCrtcInfo(dpy, res, res-crtcs[i]);
+
+ nx = (wax crtc_info-x) ? wax : crtc_info-x;
+ nw = (crtc_info-x + crtc_info-width) - nx - 2 *
c-border;
+ ny = (way crtc_info-y) ? way : crtc_info-y;
+ nh = (crtc_info-y + crtc_info-height) - ny - 2 *
c-border;
+ }
+ else { /* tile window */
+ nx = (wax last_crtc_info-x) ? wax :
last_crtc_info-x;
+ nw = (last_crtc_info-x + last_crtc_info-width) - nx -
2 * c-border;
+ ny = (way last_crtc_info-y) ? way :
last_crtc_info-y;
+ nh = (last_crtc_info-y + last_crtc_info-height) - ny
- 2 * c-border;
+
+ if(n - res-ncrtc != 0) {
+ nh = nh / (n - res-ncrtc + 1);
+ ny += (i - res-ncrtc + 1) * nh;
+ }
+ }
+ resize(c, nx, ny, nw, nh, False);
+ }
+}
