Home liberachat/#xmonad: Logs Calendar

Logs on 2023-12-30 (liberachat/#xmonad)

01:28:28 × xmonadtrack quits (~xmonadtra@user/geekosaur/bot/xmonadtrack) (Ping timeout: 276 seconds)
01:29:05 × haskellbridge quits (~haskellbr@069-135-003-034.biz.spectrum.com) (Ping timeout: 252 seconds)
01:38:52 xmonadtrack joins (~xmonadtra@069-135-003-034.biz.spectrum.com)
01:38:52 × xmonadtrack quits (~xmonadtra@069-135-003-034.biz.spectrum.com) (Changing host)
01:38:52 xmonadtrack joins (~xmonadtra@user/geekosaur/bot/xmonadtrack)
01:39:25 haskellbridge joins (~haskellbr@069-135-003-034.biz.spectrum.com)
02:20:31 SpiderPig joins (~matt@047-132-235-136.res.spectrum.com)
03:28:59 × td_ quits (~td@i53870902.versanet.de) (Ping timeout: 252 seconds)
03:30:38 td_ joins (~td@i53870904.versanet.de)
03:35:23 thunderrd joins (~thunderrd@183.182.110.131)
05:27:08 × thunderrd quits (~thunderrd@183.182.110.131) (Remote host closed the connection)
09:33:30 T_X_ joins (~T_X@diktynna.open-mesh.org)
09:33:55 × T_X quits (~T_X@diktynna.open-mesh.org) (Read error: Connection reset by peer)
11:33:18 edog joins (~edog@alcotrash.org)
12:04:00 Guest67 joins (~Guest67@2400:1a00:bd11:7857:2e0:4cff:fefc:a34d)
12:06:33 Guest67 parts (~Guest67@2400:1a00:bd11:7857:2e0:4cff:fefc:a34d) ()
13:26:21 Guest1 joins (~Guest1@2a02:8071:b80:1c20:4ed5:77ff:fea2:6ef5)
13:42:28 Guest1 is now known as Groggermaniac
13:49:18 <Groggermaniac> Hey all, currently a _NET_ACTIVE_WINDOW request with a bad window id causes (windows id) to be called, which is apparently enough to collapse context menus and such, which is a problem if opening the context menu is what caused the request in the first place: https://old.reddit.com/r/xmonad/comments/18rck0e/reaper_daw_and_xmonad/. Should I make an
13:49:19 <Groggermaniac> issue and/or a pull request for this, or is it not a bug?
13:54:17 <liskin> Hm, I guess I should have made the activateHook a MaybeManageHook instead
13:54:37 <liskin> I vaguely remember thinking about it and deciding against
13:54:40 <liskin> But why
13:55:17 <liskin> If I had you'd have the ability to ignore the request more easily
13:56:04 <liskin> Now you have to code your own event hook that filters the request before EwmhDesktops sees it
13:56:51 <liskin> Actually I think I wanted to make it possible to use PureX somehow instead of windows, but couldn't figure out a way to do it nicely
13:57:05 <liskin> Decades of legacy apis...
15:21:47 <haskellbridge> <S​olid> Looking forward to a fresh start with wlroots? :P
16:13:05 <liskin> :-)
16:32:16 <geekosaur> I already made that comment on discourse
16:34:17 T_X_ is now known as T_X
18:09:50 portnov joins (~portnov@v-20678-unlim.vpn.mgn.ru)
18:48:16 andi_ joins (~andi@user/andi/x-3467420)
18:49:15 <andi_> hello i am using xmonad and xmobar, anybody has an idea why my xmobar (config: https://paste.xinu.at/7XpAD/) looks like: https://paste.xinu.at/qOptdD/ (it's slighly shifted to the top)
18:51:42 <geekosaur> looks to me like the font's too big for the bar's size?
18:56:42 <andi_> hmm alright i'll tinker a bit
18:58:11 <geekosaur> when sizing a font you also have to consider ascenders and descenders; the descenders look like they're in the right place
19:03:21 × portnov quits (~portnov@v-20678-unlim.vpn.mgn.ru) (Remote host closed the connection)
19:37:53 portnov joins (~portnov@v-20678-unlim.vpn.mgn.ru)
19:43:47 <ectospasm> I've migrated my xmonad.hs to a new laptop, and while I was tweaking things getting things to line up correctly on the screen, I seem to have lost my DynamicWorkspaces prompt. I have (winKey, xK_h) set up to switch to the named workspace, and (winKey, xK_z) to shift the current window to the named workspace.
19:44:01 <ectospasm> The prompt isn't showing anymore, here's my current xmonad.hs: https://git.eldon.me/trey/XMonad/src/branch/master/xmonad.hs
20:08:59 <geekosaur> hm. the usual cause of a prompt not showing is a bad font, but you're using the same font everywhere and the monospace alias is mandatory
20:09:45 <geekosaur> can you check session log output (usually ~/.Xsession-errors) for any exception messages?
21:19:40 × portnov quits (~portnov@v-20678-unlim.vpn.mgn.ru) (Remote host closed the connection)
21:30:10 × Groggermaniac quits (~Guest1@2a02:8071:b80:1c20:4ed5:77ff:fea2:6ef5) (Quit: Client closed)
21:35:15 <ectospasm> geekosaur: I don't have ~/.Xsession-errors, and it looks like ~/.xmonad/xmonad.errors just has where I compiled it last (just before launching this X session with startx)
21:37:23 <geekosaur> how do you start X11?
21:37:58 <geekosaur> if you use startx, the session log is on the VT where you ran startx
21:38:33 <geekosaur> if you use sddm, the session log is ~/.local/share/sddm/xorg-session.log
21:42:22 <ectospasm> I'm not using a DM, just the Linux console. I have an alias stx='xmonad --recompile && startx', and that's how I start it.
21:42:51 <geekosaur> okay, then you need to switch back to the text VT to see the log
21:43:18 <geekosaur> ctrl+alt+F<n> (<n> is usually 1)
21:43:28 <ectospasm> Hmmmm, X is running on VT1, it doesn't spawn on VT7 like I've seen a long time ago.
21:44:06 <geekosaur> odd
21:46:07 <geekosaur> you can try switching to all the VTs to see which one is text, or `ps -ef | grep startx` and see what it shows for its tty
21:46:07 <ectospasm> This is Arch, and that's how it's been doing it for a while.
21:46:40 <ectospasm> yeah, it says tty1
21:47:17 <geekosaur> odd, that'd be VT 1
21:47:39 <ectospasm> And when I press fn + ctrl + F1 it takes me back to X; fn + ctrl + F2..F6 show login prompts on tty2-6
21:50:50 <geekosaur> wonder if they're actually switching it to graphics instead of using a different VT; unfortunately that would mean you have no session log and will have to restart X with startx output and error redirected (startx >~/.Xsession-errors 2>&1)
21:51:10 <ectospasm> I see a bunch of stuff printed in ~/.local/share/xorg/Xorg.0.log when I switch TTYs
21:51:25 <ectospasm> OK, I can add that to my alias.
21:51:33 <geekosaur> right, that's the server log
21:51:43 <geekosaur> it doesn't contain session output
21:51:49 <ectospasm> OK
21:54:11 <ectospasm> OK, now that I'm capturing .Xsession-errors I see this: user error (createFontSet)
21:54:22 <ectospasm> So it doesn't like my font?
21:54:58 <geekosaur> worse than that, it seems like your xmonad-contrib may have been built without XFT support
21:54:58 <ectospasm> That's when I press winKey + XK_z or winKey + xK_h.
21:55:14 <geekosaur> so you would have to use old-style fonts
21:56:05 <ectospasm> Hmmm, on my other machine this monospace font was working OK, also on Arch.
21:57:26 <geekosaur> how did you install xmonad and xmonad-contrib, arch packages or hackage/cabal/stack?
21:59:02 <ectospasm> Arch packages
22:00:10 <ectospasm> I know the Arch Haskell packages have always been considered broken, but I've had the least trouble with the xmonad stuff (other Haskell packages have had issues)
22:01:01 <geekosaur> odd. the dependencies indicate that it was built with -fuse_xft
22:06:03 <geekosaur> oh, wait
22:06:16 <geekosaur> the font name should be prefixed by "xft:"
22:06:38 <geekosaur> I'm not sure how it would have been working without that
22:06:56 <geekosaur> https://github.com/xmonad/xmonad-contrib/blob/master/XMonad/Prompt.hs#L325 for example
22:08:10 <geekosaur> https://github.com/geekosaur/xmonad.hs/blob/hilfy-2023/xmonad.hs#L340 is what I use
22:08:28 <ectospasm> Oh, OK. Let me try that.
22:08:42 <geekosaur> (I have a double resolution monitor so fonts are stupid big…)
22:09:30 <ectospasm> That fixed it! I don't know how it was working before without it.
22:10:19 <ectospasm> I've been mucking about with Xft.DPI a bit, when it's set to the DPI as detected by xdpyinfo fonts in GTK+ and Qt apps are too tiny to read
22:10:53 <ectospasm> I'm getting old, so anything too small is a pain to read.
22:11:22 <ectospasm> (my new laptop has an OLED screen, and it's much higher resolution than my old laptop)
22:11:39 <ectospasm> I'm regretting getting this screen, it really reduces battery life quite a bit.
22:12:09 <ectospasm> I've been told this laptop should be able to get 14 hours out of a charge, but I can only manage just under 7 with this OLED screen.

All times are in UTC on 2023-12-30.