Logs: freenode/#xmonad
| 2021-03-18 11:43:14 | <mc47> | it places it almost at the center, because that doesn't take the status bars you have into consideration |
| 2021-03-18 11:54:17 | <mc47> | What modules need better documentation, in your opinion? I want to open an issue collecting all of them in one place |
| 2021-03-18 11:54:42 | <mc47> | That's something a lot of users can contribute to, without having to be necessarily familiar with Haskell |
| 2021-03-18 11:55:25 | <mc47> | Also, it would be nice to add screenshots to the layouts. The question is, where should we host the pictures? |
| 2021-03-18 11:59:12 | → | xaltsc joins (~xaltsc@unaffiliated/xaltsc) |
| 2021-03-18 12:04:18 | <Liskni_si> | no idea about specific modules, I'd say a good tutorial about installation and basic stuff and status bars would go a long way |
| 2021-03-18 12:04:52 | <Liskni_si> | although as I said yesterday, I'm somewhat hoping that making the installation easier is better than writing docs for the current approach :-) |
| 2021-03-18 12:06:07 | <Liskni_si> | screenshots for layouts - +1, hosted either by github (they recently added it to the editor so you can just hit edit on a file and drag images) or maybe in the same place where Solid'd tutorial is going to be, or just the website repo |
| 2021-03-18 12:06:41 | <Liskni_si> | I'd tend to think that xmonad may become irrelevant sooner than github stops hosting those attachments, but we never know |
| 2021-03-18 12:11:35 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-03-18 12:25:37 | <mc47> | I have no idea about the installation, as cabal is still somewhat of a mystery to me.. but I agree that it should be simpler |
| 2021-03-18 12:25:50 | <mc47> | Yeah, github should be enough for hosting |
| 2021-03-18 12:26:34 | <geekosaur> | cabal isn't that simple, sadly. |
| 2021-03-18 12:26:53 | <geekosaur> | currently it requires a cabal file and cabal.project file. |
| 2021-03-18 12:27:28 | <geekosaur> | there's also hacky "global install" but that causes future problems especially if you have any dependencies |
| 2021-03-18 12:28:08 | <geekosaur> | basically new cabal breaks our workflow… but old cabal meant we potentially broke everyone else's (see "cabal hell") |
| 2021-03-18 12:29:45 | <mc47> | I gave up on cabal and only use stack now |
| 2021-03-18 12:30:32 | <geekosaur> | so you traded a cabal.project for a stack.yaml. I don't see much of a difference there |
| 2021-03-18 12:46:43 | <Liskni_si> | so, um, does cabal tell ghc about paths to packages in a way that prevents something like "stack exec ghc" from existing? |
| 2021-03-18 12:47:16 | <Liskni_si> | I'm fairly certain that we can get the self-recompilation work with stack-installed xmonad |
| 2021-03-18 12:47:40 | <Liskni_si> | I'm probably just going to look into it sooner or later, and deal with cabal as well |
| 2021-03-18 12:47:45 | <Liskni_si> | it can't be that hard, can it? |
| 2021-03-18 12:48:20 | <geekosaur> | cabal exec works much the same way stack exec does |
| 2021-03-18 12:48:35 | <geekosaur> | and I had an alternate workflow working but I think cabal 3.4 broke it |
| 2021-03-18 12:49:27 | <geekosaur> | there's still the question of how it finds the packages it needs, I think stack exec exposes only the resolver (which cabal has no equivalent of, so it doesn't) |
| 2021-03-18 12:50:05 | <Liskni_si> | stack exec definitely exposes all packages that would be in scope during stack build |
| 2021-03-18 12:50:15 | <Liskni_si> | otherwise my setup wouldn't work, and it's worked for many years |
| 2021-03-18 12:50:49 | × | xaltsc quits (~xaltsc@unaffiliated/xaltsc) (Ping timeout: 256 seconds) |
| 2021-03-18 12:51:12 | <geekosaur> | so you have a stack.yaml specifying packages? I don't think stack will otherwise expose anything, since the point of its existence is to not expose anything not used |
| 2021-03-18 12:51:34 | <geekosaur> | (but stack exec without a stack.yaml uses the global resolver and nothing else, I think) |
| 2021-03-18 12:51:53 | <Liskni_si> | https://github.com/liskin/dotfiles/blob/home/src-haskell/stack.yaml |
| 2021-03-18 12:52:13 | <Liskni_si> | resolver + packages that are all local git submodules |
| 2021-03-18 12:52:24 | <Liskni_si> | and then when I do stack exec, all of that is in scope |
| 2021-03-18 12:52:56 | <geekosaur> | right, the problem is that cabal has nothing equivalent to a resolver |
| 2021-03-18 12:53:21 | <Liskni_si> | I don't really care about the resolver |
| 2021-03-18 12:53:24 | <geekosaur> | the rest of stack.yaml is covered by cabal.project, but packages in scope come from the cabal file |
| 2021-03-18 12:53:37 | <geekosaur> | you implicitly do care about it because it's why stack works here |
| 2021-03-18 12:53:37 | <Liskni_si> | I just want to capture whatever was the environment during building of xmonad |
| 2021-03-18 12:53:44 | <Liskni_si> | and use that same environment to compile xmonad.hs |
| 2021-03-18 12:56:16 | <geekosaur> | https://github.com/xmonad/xmonad/issues/199 is about as close as you're getting to that but note that the same -package-env has to be specified to both the xmonad/xmonad-contrib and xmonad.hs builds |
| 2021-03-18 12:56:43 | <geekosaur> | no, ghc doesn't have this built in as such (well, it has environment files but they cause more problems than they solve) |
| 2021-03-18 12:58:46 | <Liskni_si> | https://github.com/xmonad/xmonad/issues/199#issuecomment-609433196 is indeed pretty much what I'd like to do, yeah |
| 2021-03-18 12:58:55 | <Liskni_si> | (the end of it) |
| 2021-03-18 12:59:24 | <geekosaur> | in fact that uses environment files, it just doesn't use default environment files which tend to pollute other builds unexpectedly |
| 2021-03-18 13:00:21 | <Liskni_si> | well, if I understand it correctly, we can just put "install xmonad using `cabal install -z --lib --package-env=$HOME/.xmonad/ xmonad xmonad-contrib` and everything will just work" to our homepage |
| 2021-03-18 13:00:36 | <geekosaur> | hopefully |
| 2021-03-18 13:00:50 | <Liskni_si> | maybe it's not perfect, but it sounds like quote quite an improvement from what we have now |
| 2021-03-18 13:00:59 | <geekosaur> | and until they decide to remove anything resembling v1 workflow :/ |
| 2021-03-18 13:01:00 | <Liskni_si> | s/quote // |
| 2021-03-18 13:01:17 | <Liskni_si> | oh, this uses the v1 stuff? |
| 2021-03-18 13:01:33 | <geekosaur> | it's pretending to be v1 |
| 2021-03-18 13:01:47 | <Liskni_si> | good enough I guess :-) |
| 2021-03-18 13:01:47 | <geekosaur> | that's more or less what environment files do |
| 2021-03-18 13:02:33 | <coldpress> | I find stack easier than cabal; at least upgrading with stack is easy |
| 2021-03-18 13:03:18 | <geekosaur> | cabal's not a problem once it's set up, although you need 3 files (at least if you want a local checkout of git xmonad/contrib) |
| 2021-03-18 13:03:49 | <Liskni_si> | coldpress: do you have a build script then? |
| 2021-03-18 13:04:16 | <geekosaur> | I've already upgraded xmonad-contrib once since setting this up and it Just Worked (pull, then mod-q) |
| 2021-03-18 13:09:42 | × | geekosaur quits (82650c7a@130.101.12.122) (Ping timeout: 240 seconds) |
| 2021-03-18 13:12:52 | → | xaltsc joins (~xaltsc@unaffiliated/xaltsc) |
| 2021-03-18 13:17:05 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-03-18 13:19:29 | <Liskni_si> | oh, a couple minutes of researching this and I'm reading snoyberg and hvr being mean to each other ě |
| 2021-03-18 13:19:32 | <Liskni_si> | ♥ |
| 2021-03-18 13:20:41 | <Liskni_si> | anyway, I think capturing GHC_ENVIRONMENT and GHC_PACKAGE_PATH during build (TH?) and using that when recompiling might just work; we'll just need to be somewhat careful to not do that when building distro packages |
| 2021-03-18 13:21:06 | <Liskni_si> | there's some potential for breakage but I feel this can make the installation easier for a lot of users |
| 2021-03-18 13:30:16 | → | idhugo joins (~idhugo@80-62-116-180-mobile.dk.customer.tdc.net) |
| 2021-03-18 13:48:39 | <Solid> | so I think the first draft of the tutorial is pretty much done |
| 2021-03-18 13:48:54 | <Solid> | I think it's time to open myself up to public ridicule now :) |
| 2021-03-18 13:49:39 | <Solid> | I think regardless of how the final thing will be layouted, for now I'll just post the gfmd document as a pr to the core xmonad repo; seems like the easiest way to gather suggestions |
| 2021-03-18 13:49:42 | <Solid> | any objections? |
| 2021-03-18 13:59:17 | <mc47> | awesome! I'm sure it'll be great :D |
| 2021-03-18 14:03:08 | <Solid> | I have some open questions and TODOs, but certain things I either can't answer (like if this is even comprehensible for another brain) or I'm not quite satisfied with what's there right now (e.g. because I don't use trayer I didn't have much to say here outside of what was already there) |
| 2021-03-18 14:04:36 | <Solid> | not to mention I'm not quite sure about certain things, like going heavy-ish on the xmobar configuration |
| 2021-03-18 14:05:01 | <Solid> | but I feel like this is something that a lot of people care about |
| 2021-03-18 14:06:42 | × | geekosaur quits (82650c7a@130.101.12.122) (Ping timeout: 240 seconds) |
| 2021-03-18 14:11:26 | <coldpress> | Liskni_si: yes, I use pbrisbin's build script https://brianbuccola.com/how-to-install-xmonad-and-xmobar-via-stack/ |
| 2021-03-18 14:18:07 | <Liskni_si> | coldpress: oh, cool, another stack exec ghc :-) |
| 2021-03-18 14:23:47 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-03-18 14:48:56 | × | friki quits (~friki@friki.cat) (Ping timeout: 240 seconds) |
| 2021-03-18 14:50:20 | → | friki joins (~friki@friki.cat) |
| 2021-03-18 14:52:08 | × | eqw quits (~eqw@31.134.178.99) (Ping timeout: 256 seconds) |
| 2021-03-18 14:52:44 | → | eqw joins (~eqw@31.134.178.99) |
| 2021-03-18 15:04:06 | × | coldpress quits (~coldpress@128.9.105.34.bc.googleusercontent.com) (Quit: ZNC 1.7.2 - https://znc.in) |
| 2021-03-18 15:04:23 | → | coldpress joins (~coldpress@128.9.105.34.bc.googleusercontent.com) |
| 2021-03-18 15:26:05 | → | seschwar joins (~seschwar@unaffiliated/seschwar) |
| 2021-03-18 15:33:42 | × | geekosaur quits (82650c7a@130.101.12.122) (Ping timeout: 240 seconds) |
| 2021-03-18 15:39:33 | × | xaltsc quits (~xaltsc@unaffiliated/xaltsc) (Ping timeout: 264 seconds) |
| 2021-03-18 15:49:07 | <Solid> | well there we go |
| 2021-03-18 16:00:05 | × | jeyj0[m] quits (jeyj0matri@gateway/shell/matrix.org/x-tecfsusaubfeerri) (Quit: Idle for 30+ days) |
| 2021-03-18 16:12:07 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-03-18 16:14:25 | <Solid> | mc47: re documentation: I have noted down X.L.WindowNavigation, X.A.WindowNavigation, X.P.Pass, and X.L.Spacing to be in need of a doc update |
| 2021-03-18 16:14:40 | <Solid> | oh actually X.P.Pass should be good now |
| 2021-03-18 16:15:32 | → | growpotkin joins (~growpotki@130-45-30-154.dyn.grandenetworks.net) |
| 2021-03-18 16:18:30 | <geekosaur> | when you ask if xscreensaver is still relevant, do you mean the program or the Xss extension that used to be optional in the X11 package? |
| 2021-03-18 16:18:49 | <geekosaur> | (X11 no longer has optional dependencies, they all have to be provided now) |
| 2021-03-18 16:19:01 | geekosaur | is still catching up… |
| 2021-03-18 16:19:45 | <Solid> | I mean the program |
| 2021-03-18 16:20:15 | <Solid> | as in, do people still use it? I never did but then again I never used a tray either and the seem to be quite common |
| 2021-03-18 16:20:27 | <Solid> | s/the/they/ |
| 2021-03-18 16:22:01 | <geekosaur> | I used to but haven't bothered setting it up on my current system. I have no idea what the norm is |
| 2021-03-18 16:22:57 | <geekosaur> | I think a lot of people will use things like slock, or if they're using xmonad with a DE they'll use the DE's locker |
All times are in UTC.