Logs: liberachat/#xmonad
| 2023-02-24 17:03:15 | <liskin> | either it swaps, or it forces you to move your head |
| 2023-02-24 17:03:35 | <liskin> | (and possibly also your mouse, if you're not using UpdatePointer) |
| 2023-02-24 17:05:04 | <kora9> | Solid: I think I'm starting to understand where I got confused before. I switched to a workspace on my second monitor expecting it to focus my second monitor, and instead it swapped. I suspect that what I *should've* done would be to first switch focus to the second monitor *and then* select that workspace, correct? |
| 2023-02-24 17:05:43 | <Solid> | yes, exactly |
| 2023-02-24 17:06:07 | <kora9> | Solid: Thanks. I'll give it another go, with the swap explanation in my head it's immediately making more sense |
| 2023-02-24 17:06:25 | <Solid> | you can probably set up keybindings such that M-5 to M-9 always first focuses the second monitor, and then switches to the respective workspace (and the other way around with M-1 to M-4) |
| 2023-02-24 17:06:36 | <Solid> | but not sure if that'd be even more confusing :) |
| 2023-02-24 17:06:48 | <kora9> | Yeah that might be more confusing I think |
| 2023-02-24 17:06:57 | <kora9> | I actually sort-of want to ditch workspaces entirely and just use groups |
| 2023-02-24 17:07:28 | <kora9> | I really liked using groups with cwm. I've implemented a shoddy version of it in my xmonad config, but haven't had the time to make it work well yet |
| 2023-02-24 17:08:00 | <geekosaur> | there's already DynamicWorkspaceGroups in contrib, and I think an alternative version is in a PR |
| 2023-02-24 17:08:21 | <Solid> | maybe this'll motivate liskin to upstream that one module :> |
| 2023-02-24 17:08:47 | <liskin> | what one module? |
| 2023-02-24 17:08:53 | <liskin> | do I really have anything related to this? |
| 2023-02-24 17:09:51 | <Solid> | wasn't there an issue recently in which Ivan wanted to upstream something that you wrote? |
| 2023-02-24 17:09:53 | <geekosaur> | weren't you discussing this with someone the other day in a PR? |
| 2023-02-24 17:09:56 | <geekosaur> | that, yes |
| 2023-02-24 17:10:03 | <geekosaur> | or an issue |
| 2023-02-24 17:10:08 | <Solid> | which AFAIK was related to exactly this "group" concept |
| 2023-02-24 17:10:33 | <liskin> | well it may have enabled someone to build something group-related, hm |
| 2023-02-24 17:10:40 | <liskin> | haven't actually looked into that bit |
| 2023-02-24 17:10:58 | <geekosaur> | Ivan was clearly thinking in terms of groups |
| 2023-02-24 17:11:58 | <liskin> | yes I see it now |
| 2023-02-24 17:12:24 | <liskin> | a way to attach metadata to workspaces and read/set that metadata |
| 2023-02-24 17:12:31 | <liskin> | s/metadata/a bit of data/g :-) |
| 2023-02-24 17:13:15 | <liskin> | hm, my old WorkspaceNamed module would actually benefit from it as well |
| 2023-02-24 17:13:52 | <liskin> | it carries its own patched impl of swapping workspaces so that the names are swapped as well |
| 2023-02-24 17:14:20 | <liskin> | so many things to do and so little time :-( |
| 2023-02-24 17:15:22 | geekosaur | resembles that |
| 2023-02-24 17:15:41 | <geekosaur> | although it looks like others are now carrying the torch on refiguring workspace layers |
| 2023-02-24 17:16:37 | <kora9> | Solid: Re-enabled greedyView and I'm immediately finding it more usable now that I know to expect the swaps. Now I just need to figure out how to make my second monitor xmobar show the right workspace for that monitor :) |
| 2023-02-24 17:16:39 | <geekosaur> | probably for the best, one reason I haven't been pushing forward on it is I can't see how to fix the floating "layer" with it, only how to support EWMH layers properly (or as properly as possible given the StackSet) |
| 2023-02-24 17:17:06 | <geekosaur> | I think there's already something for that? |
| 2023-02-24 17:20:39 | <mazte[m]> | Correct me if I'm under some misconception, but is there some way to resize the master pane in, say, the tall and regular tiled layouts when there's only two windows in the workspace? |
| 2023-02-24 17:21:22 | <geekosaur> | shouldn't the regular one work? |
| 2023-02-24 17:22:10 | × | kora9 quits (~kora@user/Kora9) (Quit: WeeChat 3.6) |
| 2023-02-24 17:22:37 | <mazte[m]> | I'm honestly not sure. If I use the tall or regular tiled layout, I can't resize the master pane until I add an extra window in the stack |
| 2023-02-24 17:23:01 | <mazte[m]> | I can send my config, if that'd help. The mirror tiled layout behaves the same |
| 2023-02-24 17:23:30 | <geekosaur> | hm. just worked for me but I'm using TwoPane so it kinda has to work 🙂 |
| 2023-02-24 17:23:44 | <liskin> | oh, this talk about exciting new layer developments inspired me to check opencollective and xmonad has exactly $10k in balance \o/ |
| 2023-02-24 17:24:56 | <mazte[m]> | geekosaur: If that's another layout, I'll give it a try and see whether that works |
| 2023-02-24 17:25:03 | <geekosaur[m]> | I thought we were getting pretty close to that |
| 2023-02-24 17:25:13 | <geekosaur[m]> | and yes, that's an alternative layout in -contrib |
| 2023-02-24 17:25:41 | <geekosaur[m]> | I don't use Tall or Mirror Tall |
| 2023-02-24 17:25:52 | <mazte[m]> | geekosaur[m]: Any particular reason? |
| 2023-02-24 17:26:17 | <geekosaur[m]> | not fond of the idea of splitting the slave pane as new windows are added |
| 2023-02-24 17:26:45 | <geekosaur[m]> | just personal preference, not like I'm going to hack it into core or anything like that :smile: |
| 2023-02-24 17:31:29 | <liskin> | mazte[m]: Tall is definitely meant to be resizable with just 2 windows in it |
| 2023-02-24 17:31:53 | → | ecool joins (~ecool@102.129.235.6) |
| 2023-02-24 17:31:56 | <geekosaur> | I would not expect a difference, it's still got master and slave areas so should resize |
| 2023-02-24 17:32:04 | <liskin> | ResizableTall and MouseResizableTall only add vertical resizing of slave windows, but moving the master pane divider should work in all these layouts |
| 2023-02-24 17:33:28 | <geekosaur> | added them in a spare workspace so I can test |
| 2023-02-24 17:34:20 | <mazte[m]> | liskin: In that case, the problem is definitely just how I've set it up. Using twopane, I get the same behavior |
| 2023-02-24 17:39:35 | <geekosaur> | just worked fine here, so maybe pastebin your layout |
| 2023-02-24 17:39:38 | <geekosaur> | @where paste |
| 2023-02-24 17:39:38 | <lambdabot> | Help us help you: please paste full code, input and/or output at e.g. https://paste.tomsmeding.com |
| 2023-02-24 17:40:45 | <geekosaur> | well, your whole config, not just the layout |
| 2023-02-24 17:46:41 | <mazte[m]> | gotcha |
| 2023-02-24 17:46:41 | <mazte[m]> | sec |
| 2023-02-24 17:46:56 | <mazte[m]> | i should've cleaned up by config earlier, lmao |
| 2023-02-24 17:48:46 | <mazte[m]> | https://paste.tomsmeding.com/RyGvzzaT |
| 2023-02-24 17:50:07 | <mazte[m]> | feel free to suggest/correct anything, btw. I'm more than open to criticism |
| 2023-02-24 17:50:19 | <mazte[m]> | need to clean up some of it, to say the least |
| 2023-02-24 17:53:52 | <geekosaur> | I've certainly seen worse configs |
| 2023-02-24 17:55:16 | <geekosaur> | this has me wondering if gaps or spacing is messing with messages, although one would expect them to always do so |
| 2023-02-24 17:56:54 | <mazte[m]> | geekosaur: ill take this as a win, lmao |
| 2023-02-24 17:57:40 | <mazte[m]> | geekosaur: there's nothing immediately obvious that would make it behave like that? I thought i might've just missed something, tbh |
| 2023-02-24 17:59:42 | <geekosaur> | nope, the standard bindings are still in place and they send messages to the layout, and nothing obvious is interfering with them unless it's a bug in gaps or spacing |
| 2023-02-24 18:00:08 | <geekosaur> | like, you don't have a messageControl in there that might be doing a bogus rerouting |
| 2023-02-24 18:01:43 | × | ecool quits (~ecool@102.129.235.6) (Ping timeout: 268 seconds) |
| 2023-02-24 18:01:46 | <mazte[m]> | for reference, im on 0.17.1 (NixOS), albeit this behavior is the same on arch linux and void |
| 2023-02-24 18:08:20 | <geekosaur> | hm. is it still broken if you remove the spacing stuff? |
| 2023-02-24 18:08:56 | <geekosaur> | I just popped the source to Spacing and it does some weird stuff. it's not supposed to affect this but there might be an off by one… |
| 2023-02-24 18:10:39 | <mazte[m]> | geekosaur: unfortunately |
| 2023-02-24 18:11:40 | <geekosaur> | still broken? I'm not sure that's a bad thing, since it saves us trying to figure out how to fix some downright weird stuff in the implementation of the LayoutModifier |
| 2023-02-24 18:13:10 | <mazte[m]> | additionally, even with spacing, it seems that it is trying to resize when i use the binds. What exactly do i mean by this? Well, the window borders flicker whenever any of them are resized; this happens when i try to resize with two windows, albeit none of them actually resize |
| 2023-02-24 18:13:46 | <mazte[m]> | * two windows also, albeit |
| 2023-02-24 18:15:25 | <geekosaur> | mm, that's the kind of behavior I was half expecting from Spacing |
| 2023-02-24 18:15:51 | <geekosaur> | Gaps checks out, only cares about its own messages and the rest get passed through |
| 2023-02-24 18:17:59 | → | hightower2 joins (~hightower@85.94.71.188) |
| 2023-02-24 18:19:55 | <mazte[m]> | Perhaps it would also help to mention that this behavior was present for me with a very similar config on 0.15 |
| 2023-02-24 18:20:03 | <geekosaur> | hm |
| 2023-02-24 18:22:25 | <geekosaur> | wondering if you can test that with older versions and see when it was introduced. I expect the oldest version would be the first one where `spacing` appeared, since `gaps` is ancient (it was how we reserved space for docks initially) |
| 2023-02-24 18:23:28 | → | stackdroid18 joins (14094@de1.hashbang.sh) |
| 2023-02-24 18:30:48 | <mazte[m]> | <geekosaur> "wondering if you can test that..." <- test this behavior with gaps and spacing present in a version older than 0.15.0? |
| 2023-02-24 18:31:16 | <geekosaur> | yeh |
| 2023-02-24 18:31:34 | <geekosaur> | just to see if we can nail it down that way |
| 2023-02-24 18:31:42 | <mazte[m]> | also, to avoid any misunderstanding, this is what i mean by two windows (i know it's practically obvious, but i want to make sure 100%) |
| 2023-02-24 18:31:48 | <geekosaur> | unfortunately I can't tell when Spacing was added |
| 2023-02-24 18:31:56 | mazte[m] | uploaded an image: (786KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/ujiTbtroJFjcXvNqEXhbFPJK/2023-02-24_10%3A29%3A14.png > |
| 2023-02-24 18:32:20 | <geekosaur> | I see the rewrite in 0.14 but the changelog goes back to 0.12 and then jumps to 0.9 |
| 2023-02-24 18:32:48 | <geekosaur> | yeh, that was what I figured you meant |
| 2023-02-24 18:32:55 | <mazte[m]> | geekosaur: i could perhaps set up a nix shell tomorrow to test this |
| 2023-02-24 18:33:28 | <geekosaur> | it would be very interesting to see if the 0.14 Spacing rewrite affects it |
| 2023-02-24 18:36:18 | <mazte[m]> | after removing both gaps and spacing, the behavior is the same |
| 2023-02-24 18:36:18 | <mazte[m]> | sec |
| 2023-02-24 18:36:45 | <geekosaur> | okay, this is becming really weird |
| 2023-02-24 18:37:46 | <geekosaur> | there's just no reason why you'd be seeing it with what looks to be a very simple config after removal of those two in particular, but I don't see it with a fairly complex config including putting it on a distinct workspace |
| 2023-02-24 18:38:10 | <mazte[m]> | mazte[m]: Just rebooted and recompiled too — same stuff |
| 2023-02-24 18:39:34 | <mazte[m]> | geekosaur: I was about to say that I don't think there's anything particularly fancy in there that would mess with that intended functionality |
All times are in UTC.