Logs: liberachat/#xmonad
| 2025-10-02 16:57:49 | <Enrico63> | Oh, crap, where am I unclear? :'( |
| 2025-10-02 16:57:49 | <Enrico63> | 1. My understanding is that the `IO` action I define in my plugin is run by xmobar via `async`. |
| 2025-10-02 16:57:50 | <Enrico63> | 2. The `IO` action I define in my plugin takes care of registering with DBus |
| 2025-10-02 16:57:50 | <Enrico63> | 3. Incidentally I do also spawn so process, but for doing toher stuff like creating a pipe, writing into it, but that's I believe what you refer to as fork+exec |
| 2025-10-02 16:58:01 | <Enrico63> | *spawn some processe |
| 2025-10-02 16:58:35 | <Enrico63> | Does "hard-killed" mean "no chance of clean up"? |
| 2025-10-02 16:59:44 | <geekosaur> | yes |
| 2025-10-02 17:00:25 | <Enrico63> | Then that doesn't fit with my objservations either. Because if instead of quitting xmonad I simply kill (even with -KILL) xmobar, then the dbus registration is not left hanging there |
| 2025-10-02 17:00:26 | <geekosaur> | this is forced by the OS; if the process needs to do thread cleanup, it needs to signal threads to clean up and wait for them to exit before it exits |
| 2025-10-02 17:16:57 | <geekosaur> | I'm going to have to leave soon so I can't poke at this any more right now, sorry |
| 2025-10-02 17:20:41 | <Enrico63> | No problem. I'm collecting anyway the info here https://codeberg.org/Aster89/xnobar/issues/15 so asking another day I can refer to that, if that's ok |
| 2025-10-02 17:21:04 | <Enrico63> | But thanks :) |
| 2025-10-02 17:30:33 | <liskin> | Maybe when xmobar loses its X connection it doesn't exit as cleanly as when you sigterm or sigint it? |
| 2025-10-02 17:30:57 | <liskin> | Either way, if xmobar isn't running and you still can't register on dbus, that's between you and dbus. |
| 2025-10-02 17:31:05 | <liskin> | The two of you need to agree. |
| 2025-10-02 17:31:30 | <liskin> | <geekosaur> sometimes this is just something you must deal with. https://github.com/geekosaur/xmonad.hs/blob/hilfy-2023/xmonad.hs#L448-L450 |
| 2025-10-02 17:31:35 | <liskin> | This way maybe |
| 2025-10-02 17:34:18 | <Enrico63> | "if xmobar isn't running and you still can't register on dbus". No, if xmobar isn't running, which I means I'm running my notif server as a **process**, then if I want to get in the situation where I can't register is that I register and then `kill -KILL` my notif server. But in that case, geekosaur has confirmed that's normal, as -KILL means no |
| 2025-10-02 17:34:18 | <Enrico63> | chance for clean up is given. |
| 2025-10-02 17:38:38 | × | Enrico63 quits (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) (Quit: Client closed) |
| 2025-10-02 17:39:33 | → | ft joins (~ft@p4fc2a225.dip0.t-ipconnect.de) |
| 2025-10-02 17:43:59 | → | Enrico63 joins (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) |
| 2025-10-02 17:50:35 | × | Enrico63 quits (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) (Quit: Client closed) |
| 2025-10-02 17:55:29 | → | Enrico63 joins (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) |
| 2025-10-02 17:56:56 | × | Enrico63 quits (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) (Client Quit) |
| 2025-10-02 18:41:16 | → | rascasse joins (~rascasse@user/diep) |
| 2025-10-02 19:04:23 | <rascasse> | Hi, because Im very bad with Haskell Im stuck trying to not apply window/screen spacing on the "Full" layout. I just want them for my other layouts. When I switch to Full, I want any spacing to be turned off for fullscreen experience. This is my layouts config https://b.deip.fr/p/mole-worm-shark Note for spacing Im using https://xmonad.github.io/xmonad-docs/xmonad-contrib-0.18.1.9/XMonad-Layout-Spacing.html |
| 2025-10-02 19:05:03 | <rascasse> | Can someone help me to configure that? (if it is possible) |
| 2025-10-02 19:07:53 | <rascasse> | I think I need to use setWindowSpacingEnabled False somewhere, but as I said I fail because I dont know haskell syntax |
| 2025-10-02 19:28:48 | × | Maeda quits (~Maeda@91-161-10-149.subs.proxad.net) (Quit: BRB) |
| 2025-10-02 19:47:22 | → | Maeda joins (~Maeda@91-161-10-149.subs.proxad.net) |
| 2025-10-02 19:50:04 | <Maeda> | geekosaur: Thanks for highlighting this to me. Two changes I've done and now it works. First, add spawnManage before manageHook def (i.e., `manageSpawn <> manageHook def`) and move my MymanageHook rule that defines the insertPostion at the top of the rules. All OK! Once again, your help was greatly appreciated! |
| 2025-10-02 20:35:28 | <geekosaur> | you think wrong because that's not how layouts work. generally you need to apply the spacing modifier only to the layouts you want to use it with |
| 2025-10-02 20:37:36 | <geekosaur> | rascasse, https://bin.deip.fr/upload/mole-worm-shark |
| 2025-10-02 20:40:42 | <geekosaur> | incidentally, that pastebin isa bit buggy: it backslashes $ when you edit, and when I went to remove them it backslashed them again first… |
| 2025-10-02 20:40:45 | → | Enrico63 joins (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) |
| 2025-10-02 20:41:24 | × | Enrico63 quits (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) (Client Quit) |
| 2025-10-02 20:42:34 | <rascasse> | thx! just to check then `$ spacingWithEdge 9 (tiled ||| mirror) ||| full` is correct right? |
| 2025-10-02 20:42:44 | <rascasse> | it compiles |
| 2025-10-02 20:42:52 | <rascasse> | it works! |
| 2025-10-02 20:43:03 | <rascasse> | man thanks geekosaur |
| 2025-10-02 20:43:09 | <rascasse> | many thanks* |
| 2025-10-02 20:43:21 | <rascasse> | but |
| 2025-10-02 20:44:09 | <rascasse> | for some reason when switching to Full xmobar does not draw the square icon as expected, but nothing is printed |
| 2025-10-02 20:44:36 | <rascasse> | no issue with other layers |
| 2025-10-02 20:45:47 | <rascasse> | that's my xmobar setup if it can helps https://b.deip.fr/p/ape-raven-pig |
| 2025-10-02 20:45:57 | <geekosaur> | does changing Full to Simplest help? |
| 2025-10-02 20:46:35 | <geekosaur> | (Full is a completely blank layout, which confuses some layout modifiers) |
| 2025-10-02 20:48:01 | <geekosaur> | that said, Renamed shouldn't be affected by the difference |
| 2025-10-02 20:48:10 | → | Enrico63 joins (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) |
| 2025-10-02 20:48:31 | <rascasse> | nope same issue |
| 2025-10-02 20:50:22 | <rascasse> | XMonad.Layout.Simplest right? |
| 2025-10-02 20:50:28 | <geekosaur> | right |
| 2025-10-02 20:50:58 | <geekosaur> | oh, I think I know. move the outer `renamed` so it applies only to `spacingWithEdge` |
| 2025-10-02 20:51:15 | <geekosaur> | otherwise, since the leftmost word of the full layout is the square, it's removed |
| 2025-10-02 20:52:59 | <rascasse> | renamed [CutWordsLeft 1] this one ? |
| 2025-10-02 20:53:15 | <geekosaur> | https://bin.deip.fr/upload/mole-worm-shark |
| 2025-10-02 20:53:46 | <rascasse> | my gosh Haskell >< |
| 2025-10-02 20:54:03 | <rascasse> | my small brain hurt |
| 2025-10-02 20:54:05 | <rascasse> | xD |
| 2025-10-02 20:54:17 | <geekosaur> | actually there's some unneeded parens there |
| 2025-10-02 20:54:51 | <geekosaur> | okay reload, since it's giving me the same urls |
| 2025-10-02 20:55:28 | <rascasse> | yep will try 2sec |
| 2025-10-02 20:55:29 | <geekosaur> | sorry, it backslashed my $s again, try it now |
| 2025-10-02 20:57:49 | <rascasse> | yea I spotted it |
| 2025-10-02 20:57:57 | <rascasse> | but does not compile |
| 2025-10-02 20:58:35 | <rascasse> | https://b.deip.fr/p/otter-dove-koala |
| 2025-10-02 20:58:55 | <geekosaur> | which version did you get? I had a version saved where the pastebin threw a bunch of backslashes into it that would break it |
| 2025-10-02 20:59:14 | <rascasse> | no backslashes |
| 2025-10-02 20:59:35 | <rascasse> | with `spaced = renamed [CutWordsLeft 1] $ spacingWithEdge 9` |
| 2025-10-02 20:59:53 | <rascasse> | and `$ spaced (tiled ||| mirror) ||| full` |
| 2025-10-02 21:00:29 | <geekosaur> | right, LayoutClass vs. type inference 😕 |
| 2025-10-02 21:02:19 | <rascasse> | ⚇ |
| 2025-10-02 21:03:03 | <geekosaur> | this will be ugly, I'm afraid |
| 2025-10-02 21:03:38 | <rascasse> | 🙈 |
| 2025-10-02 21:03:49 | <rascasse> | lets see |
| 2025-10-02 21:06:13 | <geekosaur> | https://bin.deip.fr/upload/mole-worm-shark |
| 2025-10-02 21:09:58 | <rascasse> | ah need to import ModifiedLayout |
| 2025-10-02 21:12:31 | <geekosaur> | also it's wrong anyway, I'm poking |
| 2025-10-02 21:12:39 | <rascasse> | ah ok |
| 2025-10-02 21:13:04 | <rascasse> | couldn't find from where to import it anyway |
| 2025-10-02 21:13:24 | <rascasse> | https://b.deip.fr/p/goose-dove |
| 2025-10-02 21:13:32 | <rascasse> | compile error |
| 2025-10-02 21:16:28 | × | Enrico63 quits (~Enrico63@2a0b:e541:10d0:0:9efc:e8ff:fe24:3213) (Quit: Client closed) |
| 2025-10-02 21:19:03 | <geekosaur> | oyes, I said it was broken |
| 2025-10-02 21:19:14 | <geekosaur> | https://bin.deip.fr/upload/mole-worm-shark less ugly than I thought and doesn't need the import |
| 2025-10-02 21:22:51 | <rascasse> | testing! |
| 2025-10-02 21:24:22 | <rascasse> | yup I confirm fixed |
| 2025-10-02 21:24:31 | <rascasse> | thanks again! |
| 2025-10-02 21:27:24 | <rascasse> | much better for this layout |
| 2025-10-02 21:28:47 | <rascasse> | btw, in Full is it possible to have the window drawn ontop of xmobar? |
| 2025-10-02 21:29:08 | <rascasse> | draw* |
| 2025-10-02 21:30:05 | <rascasse> | perhaps there is another way to toggle "fullscreen" for a given window? |
| 2025-10-02 21:30:12 | <rascasse> | I dont remember |
| 2025-10-02 21:31:14 | <geekosaur> | you need to make avoidStruts not apply to it |
| 2025-10-02 21:35:39 | <rascasse> | I see |
| 2025-10-02 21:37:09 | <rascasse> | got it `$ avoidStruts (spaced (tiled ||| mirror)) ||| full` |
| 2025-10-02 21:39:52 | <geekosaur> | I'd've just added it to spaced, but whatever |
| 2025-10-02 21:40:43 | <geekosaur> | there's also a trickier version if you want to try it at some point: https://github.com/geekosaur/xmonad.hs/blob/hilfy-2023/xmonad.hs#L183-L184 |
| 2025-10-02 21:41:21 | <geekosaur> | that is, on `Full` you use that `avoidStrutsOn []`, which means you can toggle the xmobar visible if you need to but by default it's hidden |
All times are in UTC.