You are not logged in.

#201 2010-09-05 21:31:06

litemotiv
Forum Fellow
Registered: 2008-08-01
Posts: 5,026

Re: The i3 thread

Here's a replacement for i3-wsbar and dzen2 to try out, called i3bar. It supports left/right aligning and works without extra dependencies:

http://aur.archlinux.org/packages.php?ID=40564


ᶘ ᵒᴥᵒᶅ

Offline

#202 2010-09-05 21:55:11

archman-cro
Member
From: Croatia
Registered: 2010-04-04
Posts: 943
Website

Re: The i3 thread

Since I'm a minimalist in heart and don't want to install to try (I suck), what's the equivalent to this with:

conky -c ~/.i3/conki3 | i3-wsbar -c "dzen2 -dock -h 18 -w 1024 -ta l -fn '123-9'" --show-all

Offline

#203 2010-09-06 00:16:56

Ogion
Member
From: Germany
Registered: 2007-12-11
Posts: 367

Re: The i3 thread

Hm. I just tried it out and, as long as i give it no -c argument it works fine, giving me a dockified (that is above my trayer -dock) workspace bar that correctly shows the workspaces and let's me click to switch (don't use it normally just tried for this). But, as soon as i give it a "Execute <command> to get sdtin" it dies.
If i give it a command that does not exist it still works, and says somethign about that:
zsh:1: command not found: bla
(Yes it says zsh but this is in the otuput if i3bar).

If the command is something like '/bin/echo bla' it gives this error:

Child (pid: 15867) unexpectedly exited with status 0
*** glibc detected *** i3bar: free(): invalid next size (fast): 0x09bf4ad0 ***

and then outputs a Backtrace and a Memory Map.

(I post this here because i didn't see a way to comment on the upstream page)

Ogion


(my-dotfiles)
"People willing to trade their freedom for temporary security deserve neither and will lose both." - Benjamin Franklin
"Enlightenment is man's leaving his self-caused immaturity." - Immanuel Kant

Offline

#204 2010-09-06 06:49:04

litemotiv
Forum Fellow
Registered: 2008-08-01
Posts: 5,026

Re: The i3 thread

archman-cro wrote:

Since I'm a minimalist in heart and don't want to install to try (I suck), what's the equivalent to this with:

conky -c ~/.i3/conki3 | i3-wsbar -c "dzen2 -dock -h 18 -w 1024 -ta l -fn '123-9'" --show-all

Damn you're lazy. wink

Something like

i3bar -c conky -f "-*-123-*-r-normal-*-12-*-*-*-*-60-*-*"
Ogion wrote:

Hm. I just tried it out and, as long as i give it no -c argument it works fine, giving me a dockified (that is above my trayer -dock) workspace bar that correctly shows the workspaces and let's me click to switch (don't use it normally just tried for this). But, as soon as i give it a "Execute <command> to get sdtin" it dies.

Have you tried inputting something that keeps feeding data, like conky or i3status?


ᶘ ᵒᴥᵒᶅ

Offline

#205 2010-09-06 06:57:35

Ogion
Member
From: Germany
Registered: 2007-12-11
Posts: 367

Re: The i3 thread

Yes i have. That was actually where it died.
Full output:

ogion@Gont ~ % i3bar -c conky
No Socket Path Specified, default to ~/.i3/ipc.sock
Connected to xcb
Got Atoms
Calculated Font-height: 12
Got data!
Got Subscribe Reply: {"success":true}
Got data!
Parsing Outputs-JSON...
Reconfiguring Windows...
Creating Window for output LVDS1
Destroying window for output VGA1
Got data!
Got Workspace-Data!
Conky: forked to background, pid is 15454
Got Workspace 1, name_width: 6, glyphs: 1
Got Workspace 2, name_width: 6, glyphs: 1
Got Workspace 3, name_width: 6, glyphs: 1
Got Workspace 4, name_width: 6, glyphs: 1
Got Workspace 7, name_width: 6, glyphs: 1
Drawing Bars...
Drawing Button for WS 1 at x = 0
Drawing Button for WS 2 at x = 16
Drawing Button for WS 3 at x = 32
Drawing Button for WS 4 at x = 48
Drawing Button for WS 7 at x = 64
Output VGA1 inactive, skipping...
Child (pid: 15453) unexpectedly exited with status 0
*** glibc detected *** i3bar: free(): invalid next size (fast): 0x08b66ac8 ***
======= Backtrace: =========
/lib/libc.so.6(+0x6c7b1)[0xb76717b1]
/lib/libc.so.6(+0x6d52b)[0xb767252b]
/lib/libc.so.6(cfree+0x6d)[0xb76761cd]
i3bar[0x80495de]
/usr/lib/libev.so.3(ev_invoke_pending+0x54)[0xb7771474]
/usr/lib/libev.so.3(ev_loop+0x694)[0xb7776ad4]
i3bar[0x804a1b9]
/lib/libc.so.6(__libc_start_main+0xe6)[0xb761bc76]
i3bar[0x80494e1]
======= Memory map: ========
08048000-0804d000 r-xp 00000000 08:04 3971       /usr/bin/i3bar
0804d000-0804e000 rw-p 00004000 08:04 3971       /usr/bin/i3bar
08b61000-08ba9000 rw-p 00000000 00:00 0          [heap]
b7400000-b7421000 rw-p 00000000 00:00 0 
b7421000-b7500000 ---p 00000000 00:00 0 
b75a2000-b75bd000 r-xp 00000000 08:04 7360       /usr/lib/libgcc_s.so.1
b75bd000-b75be000 rw-p 0001a000 08:04 7360       /usr/lib/libgcc_s.so.1
b75d6000-b75d7000 rw-p 00000000 00:00 0 
b75d7000-b75db000 r-xp 00000000 08:04 36859      /usr/lib/libXdmcp.so.6.0.0
b75db000-b75dc000 rw-p 00003000 08:04 36859      /usr/lib/libXdmcp.so.6.0.0
b75dc000-b75de000 r-xp 00000000 08:04 21721      /usr/lib/libXau.so.6.0.0
b75de000-b75df000 rw-p 00001000 08:04 21721      /usr/lib/libXau.so.6.0.0
b75df000-b7602000 r-xp 00000000 08:04 3056       /lib/libm-2.12.1.so
b7602000-b7603000 r--p 00022000 08:04 3056       /lib/libm-2.12.1.so
b7603000-b7604000 rw-p 00023000 08:04 3056       /lib/libm-2.12.1.so
b7604000-b7605000 rw-p 00000000 00:00 0 
b7605000-b774a000 r-xp 00000000 08:04 3030       /lib/libc-2.12.1.so
b774a000-b774c000 r--p 00145000 08:04 3030       /lib/libc-2.12.1.so
b774c000-b774d000 rw-p 00147000 08:04 3030       /lib/libc-2.12.1.so
b774d000-b7750000 rw-p 00000000 00:00 0 
b7750000-b7767000 r-xp 00000000 08:04 21736      /usr/lib/libxcb.so.1.1.0
b7767000-b7768000 rw-p 00016000 08:04 21736      /usr/lib/libxcb.so.1.1.0
b7768000-b776e000 r-xp 00000000 08:04 37816      /usr/lib/libyajl.so.1.0.9
b776e000-b776f000 rw-p 00005000 08:04 37816      /usr/lib/libyajl.so.1.0.9
b776f000-b777a000 r-xp 00000000 08:04 37177      /usr/lib/libev.so.3.0.0
b777a000-b777b000 rw-p 0000b000 08:04 37177      /usr/lib/libev.so.3.0.0
b7792000-b7794000 rw-p 00000000 00:00 0 
b7794000-b7795000 r-xp 00000000 00:00 0          [vdso]
b7795000-b77b1000 r-xp 00000000 08:04 3039       /lib/ld-2.12.1.so
b77b1000-b77b2000 r--p 0001b000 08:04 3039       /lib/ld-2.12.1.so
b77b2000-b77b3000 rw-p 0001c000 08:04 3039       /lib/ld-2.12.1.so
bfd56000-bfd77000 rw-p 00000000 00:00 0          [stack]
zsh: abort      i3bar -c conky

Ogion


(my-dotfiles)
"People willing to trade their freedom for temporary security deserve neither and will lose both." - Benjamin Franklin
"Enlightenment is man's leaving his self-caused immaturity." - Immanuel Kant

Offline

#206 2010-09-06 08:39:06

Atsutane
Trusted User (TU)
From: Germany
Registered: 2008-08-18
Posts: 96

Re: The i3 thread

Merovius is still pretty much at the beginning with the i3bar development, there will be several major changes before there'll be a real, documented release that is at the productivity level. It was a hint that one could keep an eye on that, no request to use it wide spread, at least not yet.


[blog - mostly german] - [JabberID: atsutane 0x40 freethoughts 0x2E de] - [identi.ca]

Offline

#207 2010-09-06 08:47:01

Ogion
Member
From: Germany
Registered: 2007-12-11
Posts: 367

Re: The i3 thread

Ok. It looks interesting so the above was not meant as "doesn't work moving on" tongue
(It prompted me even to look at its source a bit, i don't know any C though so that didn't help much)

Ogion


(my-dotfiles)
"People willing to trade their freedom for temporary security deserve neither and will lose both." - Benjamin Franklin
"Enlightenment is man's leaving his self-caused immaturity." - Immanuel Kant

Offline

#208 2010-09-06 12:17:56

Ogion
Member
From: Germany
Registered: 2007-12-11
Posts: 367

Re: The i3 thread

I just had the first actual bug with xcompmgr-dana. All of a sudden it ate my X output, meaning all that was left visible was the background image and teh mouse cursor. The mouse cursor behaved as normal, changing when over a terminal or the background etc. I could switch workspaces, could :wq my Firefox. Then i switched to a vt (thank you altsysrq magic), killed xcompmgr, switched back to X and tada, X was back as normal.

Ogion

Actually, i remember now why it did that. I had started xcompmgr in a terminal and i wanted it back, so i tried ctrl-z for backgrounding it. It didn't like that apparently.

Last edited by Ogion (2010-09-06 12:23:42)


(my-dotfiles)
"People willing to trade their freedom for temporary security deserve neither and will lose both." - Benjamin Franklin
"Enlightenment is man's leaving his self-caused immaturity." - Immanuel Kant

Offline

#209 2010-09-06 13:13:43

archman-cro
Member
From: Croatia
Registered: 2010-04-04
Posts: 943
Website

Re: The i3 thread

litemotiv wrote:
archman-cro wrote:

Since I'm a minimalist in heart and don't want to install to try (I suck), what's the equivalent to this with:

conky -c ~/.i3/conki3 | i3-wsbar -c "dzen2 -dock -h 18 -w 1024 -ta l -fn '123-9'" --show-all

Damn you're lazy. wink

Something like

i3bar -c conky -f "-*-123-*-r-normal-*-12-*-*-*-*-60-*-*"

This looks nice, I could remove dzen2 when i3bar gets "upgraded". Looking forward to it. smile

Offline

#210 2010-09-06 13:53:28

litemotiv
Forum Fellow
Registered: 2008-08-01
Posts: 5,026

Re: The i3 thread

Ogion, i spoke with the i3bar developer, you can send your bug report (in German wink) to: mail [at] merovius [dot] de


ᶘ ᵒᴥᵒᶅ

Offline

#211 2010-09-06 14:25:29

Ogion
Member
From: Germany
Registered: 2007-12-11
Posts: 367

Re: The i3 thread

Oh, heh ok litemotiv wink

Ogion


(my-dotfiles)
"People willing to trade their freedom for temporary security deserve neither and will lose both." - Benjamin Franklin
"Enlightenment is man's leaving his self-caused immaturity." - Immanuel Kant

Offline

#212 2010-09-08 04:57:01

Inxsible
Forum Fellow
From: Chicago
Registered: 2008-06-09
Posts: 9,079

Re: The i3 thread

litemotiv wrote:
Inxsible wrote:
gazj wrote:

I use i3 from community, and I couldn't get i3-wsbar to do anything useful!

I am in the same boat. i3-wsbar just doesn't do anything. I am using xf86-video-ati so I don't think the drivers are an issue. Maybe I should use the git version and see if it changes anything.

Update: the most likely cause is that i3-wsbar can't find the i3 socket file. In the current version the location is hardcoded to ~/.i3/ipc.sock. If you have your socket in a different location you can symlink it from .i3/ temporarily, until the -next version of wsbar becomes -main which includes a parameter for the socket location.

If you didn't have a socket running at all, you can add this config parameter:

ipc-socket ~/.i3/ipc.sock

That is probably it. I have my i3 config under ~/.config/i3

But... I am too lazy to go through the whole process again. I am currently happy with my setup of workspace bar and conky laid out on top of it. Will try out i3bar once it gets a bit more stable.

Thanks for the update litemotiv.


Forum Rules

There's no such thing as a stupid question, but there sure are a lot of inquisitive idiots !

Offline

#213 2010-09-08 08:05:46

litemotiv
Forum Fellow
Registered: 2008-08-01
Posts: 5,026

Re: The i3 thread

Inxsible wrote:

That is probably it. I have my i3 config under ~/.config/i3

But... I am too lazy to go through the whole process again. I am currently happy with my setup of workspace bar and conky laid out on top of it. Will try out i3bar once it gets a bit more stable.

For me it's really stable, the only thing i did was symlink ~/.config/i3/ipc.sock

And then removed i3-wsbar (got rid of 6 perl dependencies, yay) and dzen2. smile

My main reason to switch though was because the conky-only approach doesn't work well with dualscreen. Since i3bar docks it will always be bottom, no need to use absolute positioning.


ᶘ ᵒᴥᵒᶅ

Offline

#214 2010-09-08 09:39:24

archman-cro
Member
From: Croatia
Registered: 2010-04-04
Posts: 943
Website

Re: The i3 thread

litemotiv wrote:

And then removed i3-wsbar (got rid of 6 perl dependencies, yay) and dzen2. smile

OMG I forgot about those (and some are from AUR, so that's "nasty", since I don't use an AUR manager). Looking into i3bar! Yay!

Update: How to use a conky -c flag? I need to use some different configuration for Conky. And what's the syntax to be used in a config? Remember that the dzen2 way was to apply the commands dzen could read in conky config.

Update 2: NVM. If anyone is interested in this, just put i3bar -c 'conky -c /path/to/config'. Cheers. smile

litemotiv, I don't seem to have this 123 font in xfontsel, any ideas how to fix?

i3bar supports xft? (I kind of need that...)

Edit: AAah:

The usual elitism amongst minimal window managers: Don't be bloated, don't be fancy (simple borders are the most decoration we want to have), don't add support for Xft and blurred fonts. 

sad

Last edited by archman-cro (2010-09-09 23:02:37)

Offline

#215 2010-09-10 23:41:24

marfig
Member
From: Portugal
Registered: 2010-07-30
Posts: 176
Website

Re: The i3 thread

A poor choice of words on their part since i3bar isn't obviously a window manager, Xft fonts aren't generally blurred, and the word "bloat" is really being misused and abused; i.e. i3 already indirectly supports Xft fonts if the user chooses to use them in their terminals, only users who choose to use Xft fonts will
be "bloating" their wm, and how bars redraw means Xft has no noticeable impact on general computer performance.

It's not elitism. It's something else.
I personally find it curious that in order to have a minimalist dynamic tiling wm (which I definitely do want) that means I'm supposed to adopt an attitude towards Xft fonts, "bloat", "performance" and other senseless monikers that are today, were in the past, and will always be, hard to clearly define and impossible to measure. I appreciate minimalist dynamic tilling WMs for their usage dynamics and clear and to-the-point interfaces. Not because I have a statement I want to make to the world.

Last edited by marfig (2010-09-10 23:52:14)


I probably made this post longer than it should only because I lack the time to make it shorter.
- Paraphrased from Blaise Pascal

Offline

#216 2010-09-11 11:09:26

litemotiv
Forum Fellow
Registered: 2008-08-01
Posts: 5,026

Re: The i3 thread

marfig wrote:

A poor choice of words on their part since i3bar isn't obviously a window manager

I think that text was directly taken from the i3 goals summary: http://i3.zekjur.net/

I personally find it curious that in order to have a minimalist dynamic tiling wm (which I definitely do want) that means I'm supposed to adopt an attitude towards Xft fonts, "bloat", "performance" and other senseless monikers that are today, were in the past, and will always be, hard to clearly define and impossible to measure.

Yeah it's nonsense, it ultimately leads to a philosophical "what is function?" discussion, which can't be solved anyway with our limited resources. If a square would be the optimal functional form then nature would look like a totally different place.

archman-cro: just take any other font, i'm using terminus which actually looks better without xft.


ᶘ ᵒᴥᵒᶅ

Offline

#217 2010-09-11 11:32:01

Atsutane
Trusted User (TU)
From: Germany
Registered: 2008-08-18
Posts: 96

Re: The i3 thread

@ <sECuRE> no, xft is not supported because it requires a cairo surface or similar (AFAIR) and the main reason is that i can't get my beloved misc-fixed to be rendered exactly like with x core fonts when using xft (without antialiasing etc.)
@ <sECuRE> the lack of xcb-support was for xkb, but we use xlib for that and for x cursor theme support (we have no solution for that so far)

So it's kinda a chicken and egg problem.


[blog - mostly german] - [JabberID: atsutane 0x40 freethoughts 0x2E de] - [identi.ca]

Offline

#218 2010-09-11 11:40:05

litemotiv
Forum Fellow
Registered: 2008-08-01
Posts: 5,026

Re: The i3 thread

Atsutane wrote:

@ <sECuRE> no, xft is not supported because it requires a cairo surface or similar (AFAIR) and the main reason is that i can't get my beloved misc-fixed to be rendered exactly like with x core fonts when using xft (without antialiasing etc.)
@ <sECuRE> the lack of xcb-support was for xkb, but we use xlib for that and for x cursor theme support (we have no solution for that so far)

So it's kinda a chicken and egg problem.

No it's a subjective decision, brought forward as a functional conviction in the i3 goals. wink

Either way it's fine, it's his WM so he decides.


ᶘ ᵒᴥᵒᶅ

Offline

#219 2010-09-11 12:13:43

Atsutane
Trusted User (TU)
From: Germany
Registered: 2008-08-18
Posts: 96

Re: The i3 thread

See it from the positive side: awesome might be dropped to AUR due to the state of the XCB support from libcairo(see arch-dev-public), but i3-wm can stay in community and you don't need to build your own libcairo to use it. I may only have to move i3lock back to AUR as with 2.0 it's no longer based on slock and now needs cairo.


[blog - mostly german] - [JabberID: atsutane 0x40 freethoughts 0x2E de] - [identi.ca]

Offline

#220 2010-09-11 14:52:58

Inxsible
Forum Fellow
From: Chicago
Registered: 2008-06-09
Posts: 9,079

Re: The i3 thread

Atsutane wrote:

See it from the positive side: awesome might be dropped to AUR due to the state of the XCB support from libcairo

Tee Hee

I hate awesome, I don't know why...but I just do. wink


Forum Rules

There's no such thing as a stupid question, but there sure are a lot of inquisitive idiots !

Offline

#221 2010-09-11 18:08:51

kcirick
Member
Registered: 2010-06-21
Posts: 364

Re: The i3 thread

Inxsible wrote:

Tee Hee

I hate awesome, I don't know why...but I just do. wink

Awesome would be close 2nd of my WM of choice, only if they had tabbed/stacked layout (like Mod1+w in i3, I use it all the time), and I would only use about half of their default "widgets"

Offline

#222 2010-09-11 18:13:54

BKLive
Member
From: Georgia
Registered: 2008-01-28
Posts: 125

Re: The i3 thread

I end up using about 0 of all the widgets. Just tile the windows logically, allow me to shuffle between them, and allow me to move them out of view completely without making one screen fs and you've got me.

hadn't heard that about awesome, but it's not a biggy because I only really like awesome for the systemtray which I don't use now that I'm home and don't have to use skype (because she's sitting right here).

curiosity: does i3 still do the empty cells when you move the windows around? that was the only buggy reporting issue I had with i3. otherwise great stuff.


Main Arch Setup: HP Pavillion p7-1209, Quad-Core i3-2120 3.3Ghz, 8GB RAM, 1TB HDD, Intel Graphics
Laptop Arch Setup: Gateway lt3103u Netbook, AMD Athlon64 1.2Ghz, 2GB RAM, 250GB HDD, ATI X1270 R600

Offline

#223 2010-09-11 18:34:57

Inxsible
Forum Fellow
From: Chicago
Registered: 2008-06-09
Posts: 9,079

Re: The i3 thread

yes the empty cells still exist. The re-write is supposed to fix that, and I am not sure when the re-write will be available


Forum Rules

There's no such thing as a stupid question, but there sure are a lot of inquisitive idiots !

Offline

#224 2010-09-11 18:49:43

Atsutane
Trusted User (TU)
From: Germany
Registered: 2008-08-18
Posts: 96

Re: The i3 thread

That's not really a bug, it's just that autosnapping was never implemented. However as Inxsible already said, with the tree mapping that'll be gone.


[blog - mostly german] - [JabberID: atsutane 0x40 freethoughts 0x2E de] - [identi.ca]

Offline

#225 2010-09-11 19:23:27

marfig
Member
From: Portugal
Registered: 2010-07-30
Posts: 176
Website

Re: The i3 thread

Autosnapping isn't even desirable under the current table model. No?
I would never think branding those blank spaces as a bug; how could i3 know what window to resize between any of the ones touching the empty cell?

Where can I read more about this tree mapping?

Last edited by marfig (2010-09-11 19:23:56)


I probably made this post longer than it should only because I lack the time to make it shorter.
- Paraphrased from Blaise Pascal

Offline

Board footer

Powered by FluxBB