Home liberachat/#xmonad: Logs Calendar

Logs on 2021-06-02 (liberachat/#xmonad)

00:00:54 × tv quits (~tv@user/tv) (Ping timeout: 264 seconds)
00:01:19 tv joins (~tv@user/tv)
01:02:00 × geekosaur quits (~geekosaur@069-135-003-034.biz.spectrum.com) (Ping timeout: 245 seconds)
01:03:25 jg joins (~jrgiacone@47.149.222.98)
01:03:38 <jg> liskin how di you get around the X11-1.9.2 error
01:03:42 <jg> did
01:03:51 <jg> when updating to todays update
01:04:20 <liskin> jg: if building by stack you'll need to add X11-1.10 to extra-deps
01:05:59 <jg> ah lit, thats chill, I was on those lines but did not want to break something :) thanks
01:06:17 <jg> how was your day today ?
01:09:38 <liskin> quite productive
01:11:16 <jg> oh that's great to hear :), so good day overall?
01:11:29 <liskin> better than many other, yeah
01:11:37 <jg> im glad!
01:12:19 <liskin> but I just wasted part of the night trying to understand why Discord keeps ruining power management here
01:12:33 <liskin> so I won't get enough sleep for tomorrow :-(
01:12:54 <jg> im sorry :(, yea discord kills my laptop battery
01:13:26 <jg> I wasted the better part of last night figuring out how to update my mirrors as I kept getting timeouts
01:28:35 <liskin> oh well, I guess I'll just usbguard block-device the camera and call it a day
01:29:20 <liskin> bonus is that I don't need to worry about covering the camera ever again
01:37:38 × themainman quits (~themainma@user/themainman) (Quit: WeeChat 3.1)
01:38:08 themainman joins (~themainma@user/themainman)
01:41:03 <jg> was discord turning the camera/mic on without you asking?
01:48:08 wenzel joins (~wenzel@user/wenzel)
01:57:32 <liskin> Not really, just preventing usb power management
01:58:01 <liskin> But it probably could turn the camera on if it wanted to
01:58:27 <liskin> And since it's closed source you never know what happens
01:58:41 <liskin> Now nothing can use the camera
01:59:16 <liskin> Anyway, I guess we should mention that X11 version thing somewhere
01:59:31 <liskin> And fix a bunch of broken links on the website
02:00:19 <liskin> Bloody hell it's a lot of work :-/
02:01:03 <liskin> It's more than 5 years worth of technical debt
02:01:27 <liskin> 3am here, night, bye
02:31:30 × td_ quits (~td@94.134.91.112) (Ping timeout: 264 seconds)
02:33:10 td_ joins (~td@muedsl-82-207-238-043.citykom.de)
02:44:14 × jg quits (~jrgiacone@47.149.222.98) (Remote host closed the connection)
03:12:57 abhixec joins (~abhixec@c-67-169-139-16.hsd1.ca.comcast.net)
03:20:04 × abhixec quits (~abhixec@c-67-169-139-16.hsd1.ca.comcast.net) (Quit: leaving)
03:31:17 jg joins (~jrgiacone@47.149.222.98)
04:23:07 × jg quits (~jrgiacone@47.149.222.98) (Quit: Lost terminal)
06:01:35 <Solid> the best solution would be to not use discord :>
06:19:28 treetip is now known as globe
06:19:31 globe is now known as deebo
06:31:57 wonko joins (~wjc@62.115.229.50)
07:10:30 × bghimis quits (~bghimis@80.96.21.160) (Ping timeout: 264 seconds)
07:10:35 bghimis- joins (~bghimis@80.96.21.160)
07:11:08 cfricke joins (~cfricke@user/cfricke)
07:12:08 mc47 joins (~yecinem@89.246.239.190)
07:29:33 <mc47> liskin, is the stack.yaml in xmonad-contrib not supposed to work right now?
07:31:42 <Solid> the x11 version is definitely just a mistake
07:31:49 <Solid> but even then the xmonad version will cause problems
07:33:59 <mc47> yeah I can't build it, I tried removing the x11 from extra-deps and it still can't build
07:34:12 <mc47> which makes sense, since xmonad-0.17 doesn't exist
07:34:26 <mc47> but how are we supposed to build till we make the release?
07:35:00 <mc47> btw the activity these couple of days was quite refreshing \o/
07:35:06 <Solid> indeed :)
07:35:20 <Solid> what works for me is entering the correct x11 version (1.10) and pointing it to a local copy of xmonad $HEAD
07:37:06 <mc47> I'm a stack newbie, you're doing the latter by specifying a path and a commit?
07:37:14 <mc47> or do you just symlink xmonad in the root directory?
07:38:08 <Solid> I just specify a path (and whatever commit that repo is on will be chosen) like `- /home/slot/repos/xmonad/xmonad-slotThe`
07:39:17 <Solid> I guess this part sort of makes sense because at this point one can't build xmonad-contrib $HEAD and xmonad-0.15 together (due to the ExtensibleConfig changes)
07:39:31 <Solid> so if you're building one git version you have to build the other one as well anyways
07:44:36 gruntsplatter joins (~jason@103.2.198.78)
07:44:37 <mc47> that does make sense
07:44:44 <mc47> thanks!
07:46:18 <gruntsplatter> when I use xmonad and I move my mouse to where the gaps are my mouse jumps across the screen does anyone els have this issue with xmonad
07:46:34 <gruntsplatter> is there a fix ?
08:12:22 × themainman quits (~themainma@user/themainman) (Quit: WeeChat 3.1)
08:14:01 themainman joins (~themainma@user/themainman)
08:16:41 <Solid> oh wow there are _a lot_ of matrix nicks here now
08:16:47 <Solid> did some other channel get bridged?
08:19:47 <mc47> yes, the one that wasn't bridged
08:19:57 <mc47> the #xmonad channel
08:20:12 <Solid> that's awesome
08:20:25 <Solid> might want to add that one to the website as well then
08:36:56 × gruntsplatter quits (~jason@103.2.198.78) (Quit: WeeChat 3.1)
08:55:40 <liskin> Solid: maybe I should bridge the one Discord server I really want to be in to Matrix or something, yeah. :-/
08:57:01 <Solid> liskin: yeah that's how I've done it with one server a few friends of mine insist on using (I'm bridging directly to IRC though)
08:57:14 <Solid> the bridge itself is pretty bad, but improving it would involve me interacting with discord
08:57:17 <Solid> and also writing js
08:57:19 <Solid> so no :)
10:11:47 benin4 joins (~benin@183.82.205.186)
10:12:30 × benin quits (~benin@183.82.205.186) (Ping timeout: 272 seconds)
10:12:30 benin4 is now known as benin
10:42:38 Guest7213 joins (~Guest7213@78.32.70.245)
10:45:07 × berberman quits (~berberman@user/berberman) (Ping timeout: 265 seconds)
10:45:13 berberman_ joins (~berberman@user/berberman)
11:43:21 hpd[m] parts (~hpdhpdeif@2001:470:69fc:105::230) ()
11:57:48 MrElendig joins (~Urist@user/mrelendig)
12:16:41 Guest73 joins (~Guest73@124.40.244.150)
12:18:06 benin8 joins (~benin@106.198.94.140)
12:18:15 <Guest73> When using xmonad-git how can I look at documentation? Only in source or some place on hackage?
12:19:55 × benin quits (~benin@183.82.205.186) (Ping timeout: 245 seconds)
12:21:43 benin joins (benin@gateway/vpn/protonvpn/benin)
12:23:41 × benin8 quits (~benin@106.198.94.140) (Ping timeout: 264 seconds)
12:27:49 <Guest73> Any1
12:28:48 <liskin> Guest73: you can build the html docs locally via cabal haddock or stack haddock, but it's not available anywhere on the web right now
12:29:08 <liskin> it's somewhere in my todo list though
12:30:30 <Guest73> Thanks. I will stick to code. No idea about cabal and stack. Not a Haskell programmer. Just a xmonad user.
12:32:33 <liskin> Oh you're installing from distro? Might be a good idea to ask the maintainers to make a doc package as well.
12:32:59 <liskin> In Debian, there is a docs package.
12:33:06 × benin quits (benin@gateway/vpn/protonvpn/benin) (Ping timeout: 272 seconds)
12:33:08 <liskin> (But there's no -git package :-/)
12:33:36 <liskin> Possibly there already is a doc package for your distro?
12:36:36 × Guest73 quits (~Guest73@124.40.244.150) (Ping timeout: 250 seconds)
12:47:04 benin joins (~benin@183.82.205.186)
12:55:06 <Solid> layouts are so weird
12:56:37 <Solid> liskin: everything you said makes sense and yet the current solution for #281 still works, even with decorated layouts
12:57:13 <Solid> I think if the check the description on both sides every time we can avoid not hiding the layouts though
13:02:47 <liskin> that's really weird :-/
13:05:51 <liskin> I think I might write a prop test or something
13:06:45 <liskin> otherwise I'd have to rebase everything onto current masters and that might be more trouble, as I have some custom inspection instances for NewSelect and shit
13:28:07 <liskin> but then I'd be tempted to clean up the tests and use hspec, so I'll just write a new xmonad config instead
13:32:33 <Solid> hah
13:33:25 × themainman quits (~themainma@user/themainman) (Read error: Connection reset by peer)
13:34:00 themainman joins (~themainma@user/themainman)
13:42:46 × benin quits (~benin@183.82.205.186) (Ping timeout: 272 seconds)
13:45:14 <liskin> okay I'm not crazy
13:45:33 <liskin> I can reproduce dangling deco with main = xmonad bluetileConfig
13:47:28 <liskin> open two terminals, mod-f to switch to fullscreen, mod-d to switch to tile2, now through the gap they can be seen and stay there after closing both terminals
13:48:05 <liskin> that is, at 246600f70d6365c612ea9bfd31efd50c50a6c861, not at 0084a1af5c08fa3015af5aa2fd45d0c165d86a67
14:07:36 <Solid> huh interesting
14:07:41 <Solid> okay so that clears that up
14:10:00 benin joins (~benin@183.82.205.186)
14:11:03 geekosaur joins (~geekosaur@069-135-003-034.biz.spectrum.com)
14:11:47 <liskin> Solid: I pushed a fixup
14:11:47 <Solid> does 0084a1af5c08fa3015af5aa2fd45d0c165d86a67 fix this though? because it _should_ but I'm not sure of anything anymore right now :/
14:12:01 <liskin> 0084a1af5c08fa3015af5aa2fd45d0c165d86a67 does fix it indeed
14:12:34 <liskin> as I said, whenever you flip CL/CR outside of choose, some Hide messages might not get sent
14:12:47 <liskin> or some extra Hides may get sent
14:12:54 <liskin> which is what the fixup deals with
14:13:17 <Solid> liskin: would that fixup not run into the same problems regarding having (CL, CL) and (CR, CR) as the first version?
14:13:22 <liskin> I still don't understand what those flips were meant to do
14:13:40 <liskin> it's not (CL, CL) et al that is the problem
14:14:06 <liskin> the problem is when you switch to another layout without sending Hide to the old one
14:14:34 <liskin> the fixup just stops a possible flicker when JumpToLayout is asked to switch to the current layout
14:16:27 <Solid> ooooh okay
14:17:17 <liskin> actually, I haven't tested the fixup
14:17:32 <liskin> it's entirely possible it breaks it entirely
14:18:28 <liskin> oh no it doesn't :-)
14:18:46 <Solid> yeah can confirm it also works for me
14:19:39 <liskin> anyway what I had in mind that the JumpToLayout handler can very likely be simplified further, if we're willing to risk calling handleMessage on both sides at all times
14:27:29 <liskin> also, if the tree of layouts isn't exactly the right fold of a list, jumping to the layouts in the right subtree doesn't work :-/
14:28:42 <liskin> with https://store.lisk.in/tmp/2021-06-02-152751_704x446_scrot.png I can't switch to Fullscreen
14:33:28 <Solid> huh interesting
14:33:33 <Solid> this this work originally?
14:33:39 <Solid> s/this/did/
14:34:36 <liskin> yeah I think before 246600f70d6365c612ea9bfd31efd50c50a6c861 handling of JumpToLayout would return Nothing when the layout wasn't there
14:34:46 <liskin> and the logic relies on that
14:35:46 <liskin> it's a bit weird though, hm, choose also returns Nothing sometimes
14:36:48 seschwar joins (~seschwar@user/seschwar)
14:37:17 <liskin> but those nested cases might force a switch of CL to CR or the other way around even if the layout wasn't found, apparently
14:41:50 <liskin> Solid: https://github.com/xmonad/xmonad/pull/281/files/0084a1af5c08fa3015af5aa2fd45d0c165d86a67..28ac3178ae0a6959e40d699fd764d0cf53b635c0
14:41:57 <liskin> oh not this one
14:42:06 <liskin> this: https://github.com/xmonad/xmonad/pull/281/commits/3e05edfaf7dd5800ed0affcde2f7a400aa4e8abe
14:42:42 <liskin> but we should probably really write the prop test after all, this is too brittle
14:44:14 <Solid> liskin: oh that's really neat
14:44:56 <Solid> is there infrastructure for message handling tests?
14:45:28 <liskin> do you need more than just a call to handleMessage?
14:46:12 <Solid> mh... probably not
14:46:17 <liskin> the only possibly difficult bit is that type T uses Int as layout, but here you'll need something with both description and state to check that Hide was delivered
14:46:31 <liskin> might just do a bit of bit arithmetic though
14:56:07 bghimis- parts (~bghimis@80.96.21.160) (WeeChat 3.1)
14:57:48 × cfricke quits (~cfricke@user/cfricke) (Quit: WeeChat 3.1)
15:32:19 benin5 joins (benin@gateway/vpn/protonvpn/benin)
15:34:30 × benin quits (~benin@183.82.205.186) (Ping timeout: 264 seconds)
15:34:31 benin5 is now known as benin
15:45:15 benin1 joins (~benin@183.82.205.186)
15:47:00 × benin quits (benin@gateway/vpn/protonvpn/benin) (Ping timeout: 245 seconds)
15:47:01 benin1 is now known as benin
18:26:44 tremon joins (~tremon@217-63-61-89.cable.dynamic.v4.ziggo.nl)
18:46:16 × wenzel quits (~wenzel@user/wenzel) (Read error: Connection reset by peer)
18:46:36 wenzel joins (~wenzel@user/wenzel)
18:58:49 × wenzel quits (~wenzel@user/wenzel) (Quit: Leaving)
19:39:31 Guest73 joins (~Guest73@124.40.244.150)
19:41:29 <Guest73> Is there a way to add description to key bindings? I am thinking of generating help automatically.
19:43:35 <liskin> Guest73: reminds me of https://old.reddit.com/r/xmonad/comments/nhu6uu/keybindings_generator/
19:43:55 <liskin> also https://hackage.haskell.org/package/xmonad-contrib-0.16/docs/XMonad-Util-NamedActions.html
19:52:51 <Guest73> I liked this part: An additional step may be to have them displayed in a picture like those in the xmonad documentation
19:56:22 <liskin> we have an open issue that mentions something similar: https://github.com/xmonad/xmonad/issues/287
19:56:51 <liskin> (and I can almost guarantee I will not work on this myself)
19:59:28 × Guest73 quits (~Guest73@124.40.244.150) (Ping timeout: 250 seconds)
20:00:19 Guest73 joins (~Guest73@124.40.244.150)
20:03:20 × Guest73 quits (~Guest73@124.40.244.150) (Client Quit)
20:11:06 benin0 joins (benin@gateway/vpn/protonvpn/benin)
20:13:30 × benin quits (~benin@183.82.205.186) (Ping timeout: 264 seconds)
20:13:31 benin0 is now known as benin
20:43:42 lyiriyah[m] joins (~lyiriyahm@2001:470:69fc:105::cc0)
20:51:58 <lyiriyah[m]> Hi -- trying to figure out how to integrate IndependentScreens and PerWorkspace. IndependentScreens documentation doesn't really tell me anything. If anyone could should some insight that'd be great
20:52:54 lyiriyah52 joins (~lyiriyah@host109-157-157-23.range109-157.btcentralplus.com)
20:53:02 lyiriyah52 parts (~lyiriyah@host109-157-157-23.range109-157.btcentralplus.com) ()
20:54:05 × benin quits (benin@gateway/vpn/protonvpn/benin) (Ping timeout: 245 seconds)
21:01:58 <lyiriyah[m]> * Hi -- trying to figure out how to integrate IndependentScreens and PerWorkspace. IndependentScreens documentation doesn't really tell me anything. If anyone could share some insight that'd be great
21:15:23 × mc47 quits (~yecinem@89.246.239.190) (Remote host closed the connection)
21:19:59 × Guest7213 quits (~Guest7213@78.32.70.245) (Quit: Client closed)
21:26:50 × geekosaur quits (~geekosaur@069-135-003-034.biz.spectrum.com) (Remote host closed the connection)
21:27:00 <liskin> lyiriyah[m]: you can use https://hackage.haskell.org/package/xmonad-contrib-0.16/docs/XMonad-Layout-IndependentScreens.html#v:marshall to obtain a WorkspaceId to use in PerWorkspace
21:28:24 <lyiriyah[m]> <liskin "lyiriyah: you can use https://ha"> Ok, but what is ScreenId?
21:29:10 geekosaur joins (~geekosaur@069-135-003-034.biz.spectrum.com)
21:29:18 <liskin> lyiriyah[m]: you can just use a number, 0 being the primary screen, 1 being the next (in whatever order xinerama reports it to xmonad)
21:29:27 <lyiriyah[m]> Oh I see
21:29:30 <lyiriyah[m]> Thanks
21:30:52 <lyiriyah[m]> How would I marshall more than one workspace?
21:34:49 <liskin> *Main XMonad.Layout.IndependentScreens> map (marshall 1) ["one", "two"]
21:34:51 <liskin> ["1_one","1_two"]
21:35:17 <lyiriyah[m]> Ah, thanks so much. I'll give it a go
21:41:17 × tremon quits (~tremon@217-63-61-89.cable.dynamic.v4.ziggo.nl) (Quit: getting boxed in)
21:50:24 habib joins (~habib@89.38.69.198)
21:50:35 × habib quits (~habib@89.38.69.198) (Quit: habib)
22:09:36 × seschwar quits (~seschwar@user/seschwar) (Quit: :wq)
22:11:30 <liskin> geekosaur: what has to come from somewhere?
22:12:07 <liskin> (we can continue the discussion there, just didn't want to add unnecessary noise by asking for this clarification)
22:12:16 <geekosaur> GHC_ENVIRONMENT and friends
22:12:34 <liskin> that would come from the build of xmonad library itself
22:12:37 <geekosaur> you have to have somewhere to capture them from, so at least once have to build with cabal or stack
22:13:08 <liskin> I kind of assumed cabal/stack set them when invoking ghc to build
22:13:11 <geekosaur> oh. that's not certain to be the same as the one we need, though, especially if there are added deps
22:13:26 <liskin> yes, that is a very legitimate concern
22:14:37 × de-vri-es quits (~de-vri-es@voyager.de-vri.es) (Quit: Bathrobe!)
22:15:05 de-vri-es joins (~de-vri-es@voyager.de-vri.es)
22:22:16 <geekosaur> come to think of it, when cabal or stack is building xmonad-contrib, it's doing so under /tmp
22:22:35 <geekosaur> so the environment you capture potentially won't include it or xmonad
22:23:30 <liskin> I thought it'd point somewhere in ~/.stack/snapshots or ./.stack-work/…
22:23:53 <liskin> anyway, I'm leaning towards not even trying this
22:24:22 <geekosaur> it may point under there but it's a PATH-like variable so it may point separately to a bunch of different package dirs
22:24:23 <liskin> if cabal --package-env ~/.xmonad/ works, then cabal is a non-issue
22:24:38 <geekosaur> only way to know is to try it and then hope they don''t change it
22:24:50 <geekosaur> oh, I know that works, it was the first thing I tried
22:25:05 <geekosaur> think I erased my setup afterward though
22:25:26 <liskin> and for stack I think a more explicit approach like configuring the stack.yaml path or detecting stack.yaml at known locations seems like a better approach
22:25:28 <geekosaur> yeh, it's gone now
22:25:37 <liskin> I'm a bit confused about --package-env though
22:25:45 <liskin> it sounds like such a perfect solutions
22:25:51 <liskin> why isn't it recommended everywhere?
22:26:30 <geekosaur> because most people work in terms of projects, not quasi-global installs
22:26:58 <geekosaur> --package-env is what replaces global installs with cabal; for stack it's the global environment that replaces it
22:27:22 <liskin> are you sure most people do that?
22:27:40 <geekosaur> reasonably certain, yes
22:27:43 <liskin> I thought most people just have a single xmonad.hs
22:28:18 <geekosaur> oh, I was talkingg in general. I have no idea what most xmonad users do at this point
22:28:21 <liskin> I mean, I consider myself an advanced xmonad "user" :-), and I had a single xmonad.hs until this year
22:28:28 <geekosaur> exceot get confused at ancient instructions :)
22:28:45 <liskin> oh
22:29:00 <liskin> I meant why isn't it recommended for installing xmonad
22:29:08 <geekosaur> actually I think most xmonad users install via system package manager and the really global env is still there
22:29:44 <liskin> is it just that we haven't updated our instructions since cabal went v2 and 3rd party guides are a mess in general?
22:30:26 <liskin> because if cabal install --package-env ~/.xmonad/ works just as well as the old v1 cabal install, then we should just recommend that
22:30:38 <geekosaur> pretty much
22:30:42 <liskin> oh, good point about package managers
22:31:07 <liskin> arch people can probably even get the git version via their package manager
22:31:26 <liskin> and as long as they manage to not run into dynamic linking breakage, they might live in blissful ignorance of cabal/stack
22:32:06 <liskin> come to think of it, I should probably be qualified enough to provide builds of git xmonad for ubuntu and debian
22:32:20 <liskin> brilliant, more things to put on my todo list :-D
22:34:08 <geekosaur> evidently they patch xmonad to include -dynamic in its build command
22:35:14 × wonko quits (~wjc@62.115.229.50) (Quit: See You Space Cowboy..)
22:36:54 <liskin> understandably
23:00:41 benin0 joins (~benin@183.82.205.186)
23:00:54 dmwit joins (~dmwit@pool-96-255-233-247.washdc.fios.verizon.net)

All times are in UTC on 2021-06-02.