Logs on 2021-10-24 (liberachat/#xmonad)
| 00:16:52 | → | mariari joins (~mariari@user/mariari) |
| 00:21:26 | <mariari> | has anyone experienced xmonad locking up horribly when a program demands constant focus? I've notice with some wine games and other applications that this could happen, causing xmonad to spasm shifting the boarder color of the application between the focus and unfcous state. When the application stops, these inputs are buffered and the machine becomes unusable for a while after this |
| 00:27:47 | <L29Ah> | mariari: no |
| 00:27:53 | <L29Ah> | but i use wine's virtual desktop feature |
| 00:28:02 | <L29Ah> | 10/10 do recommend |
| 00:29:16 | <mariari> | It's not just wine sadly, it seems any application that seems to demand attention in a certain way it freaks xmonad out, I can't find any bug reports of a similar problem |
| 00:29:36 | <L29Ah> | can you name a few? |
| 00:29:44 | <mariari> | Games mostly |
| 00:29:57 | <L29Ah> | don't see that with xonotic |
| 00:30:01 | <mariari> | not all games |
| 00:30:04 | <mariari> | xonotic does not have the issue |
| 00:30:14 | <mariari> | it's only certain applications that demand the screen attention in certain ways |
| 00:30:58 | <mariari> | it's with how certain games do a call, most games do it in a way that does not freak out xmonad |
| 00:42:31 | <geekosaur> | this may be improved with the new ewmh (which is still being ironed out), we don't currently advertise that we handle focus changes correctly and this confuses a number of applications |
| 00:42:47 | <mariari> | ahh is that what may be causing this? |
| 00:43:25 | <geekosaur> | some of the apps that do it have options to force old-style focus changing, and they behave correctly in that case. (I think vlc is one?) |
| 00:45:58 | <geekosaur> | so hopefully when we get 0.17 out (should be soon) this should be mostly fixed. if it isn't, open an issue then |
| 00:46:02 | × | mc47 quits (~mc47@xmonad/TheMC47) (Remote host closed the connection) |
| 00:46:48 | <mariari> | OK, do you know the rough behavior I'm alluding to? I just recorded an application that does it |
| 00:46:55 | <geekosaur> | (we're targeting before the end of the month. how "before" remains to be seen) |
| 00:46:59 | <mariari> | it makes xmonad fairly unusable |
| 00:47:07 | <mariari> | ahh nice, I hope it ends up on the stack lts soon, as that's how I use xmonad |
| 00:47:23 | <geekosaur> | the focus flickering thing is fairly well known, yes |
| 00:47:46 | <geekosaur> | it's a largeish part of why the ewmh revamp that is still in progress |
| 00:51:38 | <geekosaur> | …and there may still be breakage afterward from apps that ignored the part of ewmh that says that a window manager is not required to obey application-initiated focus changes, but we can hope those are few (and default will be to honor them, it'll be configurable) |
| 01:01:00 | <mariari> | do you know the specific name for the event that causes this? |
| 01:06:20 | <geekosaur> | _NET_ACTIVE_WINDOW |
| 01:08:28 | <geekosaur> | hm, actually I'm not exactly sure we handle it if an app just calls XSetInputFocus directly, we may just slap it back to what we think it should be |
| 01:08:44 | <geekosaur> | apps are not supposed to call that directly except specifically to their own subwindows |
| 01:09:24 | <geekosaur> | which is why _NET_ACTIVE_WINDOW exists, to allow apps to request that the wm change focus |
| 01:19:46 | × | benin quits (~benin@183.82.207.116) (Ping timeout: 244 seconds) |
| 01:20:48 | <mariari> | I see, so if that gets called often hence the boarder flickering and xmonad lag/unresposiveness |
| 01:22:12 | <geekosaur> | well, some apps just get uppity if they change focus and then something else (like the window manager) changes it somewhere else, so they change it again |
| 01:22:20 | <geekosaur> | and you end up with an ugly loop |
| 01:23:27 | → | benin joins (~benin@183.82.207.116) |
| 01:23:37 | <mariari> | but what if it's just 1 application stealing focus? |
| 01:23:42 | <mariari> | that is what I have at least |
| 01:23:52 | <mariari> | probably calling it constantly so it flickers on 1 window |
| 01:24:29 | <geekosaur> | right, 1 app changes focus to itself, xmonad changes it back, app changes it bvack to itself, lather rinse repeat |
| 01:24:41 | <mariari> | changes it back to where exactly? |
| 01:24:48 | <mariari> | that is the confusing part for me |
| 01:24:57 | <mariari> | as you can have the item focused, and then the next moment it works like that |
| 01:25:03 | <mariari> | so I'm unsure of where the focus is moving to |
| 01:25:12 | <geekosaur> | to whatever window had focus before the app forcibly changed it |
| 01:25:45 | <geekosaur> | if this is happening while the app has focus then you're seeing the reverse form iof it, some app is stealing focus and then xmonad sets it back to your app |
| 01:25:55 | <geekosaur> | you still get the loop |
| 01:26:22 | <geekosaur> | alternately you can have two apps fighting over which one shoyuld get focxus and stealing it from each other |
| 01:26:23 | <mariari> | well I have an example where in a menu the game does not steal focus, so the game is focused, then when I close the windows it acts like that |
| 01:26:26 | <geekosaur> | *focus |
| 01:26:32 | <mariari> | so it's one app steal focus from itself it seems |
| 01:27:46 | <geekosaur> | hm. either (1) this is the not-fully-supporting ewmh thing, or (2) we've once seen an app that behaved or not depending on whether a compositor was running or not |
| 01:27:56 | <geekosaur> | notably that one also involved menus |
| 01:28:03 | <geekosaur> | (and it was a game design app) |
| 01:28:54 | <mariari> | huh a compositor could effect it, I could perhaps put on whatever compoton is called these days to see if that makes a difference |
| 01:28:59 | <mariari> | I think it's likely 1) |
| 01:29:02 | <geekosaur> | picom, apparently |
| 01:29:16 | <mariari> | as I've launched the game on stumpwm, and it didn't freak out but it stole attention every time I tried to go away |
| 01:29:37 | <geekosaur> | although I just run compton because (a) that's what's packaged on ubuntu (b) I keep hearing about performance problems with picom |
| 01:31:03 | <geekosaur> | ok, so that's actually the same issue but stumpwm apparently handles it better than we do, or maybe that's because of our not advertising _NET_ACTIVE_WINDOW |
| 01:31:49 | <geekosaur> | again I'd be interested in hearing of the ewmh fixes in 0.17 help or even just change the behavior |
| 01:32:04 | <geekosaur> | *if the |
| 01:32:07 | <mariari> | yeah even it just making it not make xmonad unusable if it lasts for any length of time is great |
| 01:32:20 | <mariari> | yesterday I had it switch between workspace 2 and 3 for a solid 10 minutes before I killed my entire x11 |
| 01:32:33 | <geekosaur> | they're not complete yet so I can't even point you at a PR to install locally |
| 01:38:43 | <mariari> | yeah that's fine, hopefully it'll be on an lts eventually |
| 01:40:02 | <mariari> | with this particular application I hope to get it's call fixed, so it'll be hard to tell if the game fixes it as I can't save an older version sadly |
| 01:41:17 | <mariari> | I wish I had better examples for this, I only do for the other bugs I face |
| 01:43:38 | <mariari> | there is a fun one with windows that close bringing focus to the last window in the stack, so in a sub layout with tabs where one has a tab in the last slot in the stack it moves to the non visible one |
| 01:50:14 | × | catman quits (~catman@user/catman) (Ping timeout: 265 seconds) |
| 01:51:54 | → | catman joins (~catman@user/catman) |
| 02:03:10 | × | banc quits (banc@gateway/vpn/airvpn/banc) (Ping timeout: 244 seconds) |
| 02:22:09 | → | banc joins (banc@gateway/vpn/airvpn/banc) |
| 02:38:49 | × | td_ quits (~td@muedsl-82-207-238-090.citykom.de) (Ping timeout: 244 seconds) |
| 02:40:49 | → | td_ joins (~td@muedsl-82-207-238-060.citykom.de) |
| 03:21:40 | × | mvk quits (~mvk@2607:fea8:5cc1:300::4b63) (Ping timeout: 252 seconds) |
| 06:03:42 | → | benin5 joins (~benin@183.82.207.116) |
| 06:06:05 | × | benin quits (~benin@183.82.207.116) (Ping timeout: 260 seconds) |
| 06:06:05 | benin5 | is now known as benin |
| 07:49:14 | → | ircbrowse_tom joins (~ircbrowse@2a01:4f8:1c1c:9319::1) |
| 07:49:19 | Server | sets mode +cnt |
| 07:56:43 | × | darkstarx quits (~darkstard@2601:1c2:300:c8a0::6c32) (Remote host closed the connection) |
| 07:57:03 | → | darkstarx joins (~darkstard@2601:1c2:300:c8a0::6c32) |
| 09:05:31 | <liskin> | hm, this endless flicker issue was historically known as https://code.google.com/archive/p/xmonad/issues/200 and I haven't seen it in years… |
| 09:06:32 | <liskin> | mariari, geekosaur: it's also worth noting that all the changes related to EWMH that we wanted to do before the 0.17 release are already in |
| 09:07:12 | <liskin> | what is left is just a revamp of the user/configuration interface of the EWMH modules, but functionally we're good already |
| 09:07:59 | <liskin> | Solid: did we decide what to do with the X.L.Fullscreen deprecation warning? |
| 09:24:04 | → | seschwar joins (~seschwar@user/seschwar) |
| 10:37:07 | × | humky quits (~humky@user/humky) (Quit: Leaving) |
| 11:57:04 | → | eqw joins (~eqw@31.134.178.99) |
| 12:44:15 | → | mc47 joins (~mc47@xmonad/TheMC47) |
| 13:37:52 | <Solid> | liskin: we did not and I still don't quite know how to feel about it |
| 13:38:43 | <Solid> | btw, did you self assign 341 because you don't want anyone else to interfer or because you'll do it if no one else does? :) |
| 13:55:24 | → | rekahsoft joins (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) |
| 13:55:32 | × | catman quits (~catman@user/catman) (Remote host closed the connection) |
| 13:58:00 | × | rekahsoft quits (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) (Remote host closed the connection) |
| 14:12:04 | → | rekahsoft joins (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) |
| 14:23:38 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Quit: Leaving) |
| 14:26:43 | → | catman joins (~catman@user/catman) |
| 14:34:51 | <liskin> | Dunno what's 341 (on a walk) but definitely the latter |
| 14:34:58 | <liskin> | Solid: ^ |
| 14:35:58 | × | rekahsoft quits (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) (Remote host closed the connection) |
| 14:36:04 | <liskin> | I'm working on packdeps today, then then announcement and/or ppPrinters |
| 14:36:06 | <Solid> | aight, thanks! (it's the "update contributing" issue) |
| 14:37:37 | <liskin> | Oh, then that's a bit of both. I have some thoughts that I wanted to get out in writing but that doesn't mean I need to do the whole revamp |
| 14:48:16 | <Solid> | aight, I'll do some of the menial things then (like deleting config and migrating style) |
| 14:50:37 | → | benin8 joins (benin@gateway/vpn/protonvpn/benin) |
| 14:52:50 | × | benin quits (~benin@183.82.207.116) (Ping timeout: 260 seconds) |
| 14:52:51 | benin8 | is now known as benin |
| 15:10:43 | <liskin> | Thx |
| 15:12:05 | <liskin> | I mean, feel free to do it all if you want, I'll just make sure it says what I want it to say |
| 15:14:35 | <Solid> | sure, I'll do as much as I can before PnP later this evening |
| 15:18:49 | → | benin8 joins (benin@gateway/vpn/protonvpn/benin) |
| 15:21:50 | × | benin quits (benin@gateway/vpn/protonvpn/benin) (Ping timeout: 258 seconds) |
| 15:21:51 | benin8 | is now known as benin |
| 15:34:37 | <liskin> | What's PnP? |
| 15:38:14 | → | benin5 joins (~benin@183.82.207.116) |
| 15:38:56 | × | benin quits (benin@gateway/vpn/protonvpn/benin) (Remote host closed the connection) |
| 15:38:56 | benin5 | is now known as benin |
| 15:42:01 | <Solid> | pen and paper (think DnD, but a different system (we play call of cthulu)) |
| 15:43:59 | <liskin> | I see |
| 16:06:40 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 16:07:03 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 16:12:52 | → | benin4 joins (~benin@106.198.88.230) |
| 16:15:07 | × | benin quits (~benin@183.82.207.116) (Ping timeout: 258 seconds) |
| 16:15:08 | benin4 | is now known as benin |
| 16:20:11 | × | benin quits (~benin@106.198.88.230) (Read error: Connection reset by peer) |
| 16:22:04 | → | benin joins (~benin@183.82.207.116) |
| 16:22:28 | → | qbt joins (~qbt@user/edun) |
| 18:44:05 | → | humky joins (~humky@user/humky) |
| 19:17:00 | → | rekahsoft joins (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) |
| 19:24:12 | × | qbt quits (~qbt@user/edun) (Quit: Leaving.) |
| 20:19:17 | × | rekahsoft quits (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) (Remote host closed the connection) |
| 20:20:13 | × | humky quits (~humky@user/humky) (Quit: Leaving) |
| 20:21:19 | → | rekahsoft joins (~rekahsoft@cpe0008a20f982f-cm64777d666260.cpe.net.cable.rogers.com) |
| 20:36:26 | → | seiryn joins (~seiryn@2a01cb040147e000baac7520288c550c.ipv6.abo.wanadoo.fr) |
| 20:40:12 | × | catman quits (~catman@user/catman) (Ping timeout: 244 seconds) |
| 20:40:20 | → | catman_ joins (~catman@user/catman) |
| 20:40:52 | catman_ | is now known as catman |
| 20:40:52 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 20:47:11 | × | seiryn quits (~seiryn@2a01cb040147e000baac7520288c550c.ipv6.abo.wanadoo.fr) (Quit: WeeChat 3.3) |
| 20:48:17 | → | seiryn joins (~seiryn@2a01cb040147e000baac7520288c550c.ipv6.abo.wanadoo.fr) |
| 21:09:54 | × | seiryn quits (~seiryn@2a01cb040147e000baac7520288c550c.ipv6.abo.wanadoo.fr) (Quit: WeeChat 3.3) |
| 21:14:37 | <jakefromstatefar> | Whenever my mumble client's tiled, if it gets resized, it unmaps, and is incapable of re-showing itself. |
| 21:14:48 | <jakefromstatefar> | The same thing happens if I switch workspaces. |
| 21:15:01 | <jakefromstatefar> | Does anyone know why? |
| 21:19:04 | × | mc47 quits (~mc47@xmonad/TheMC47) (Remote host closed the connection) |
| 21:24:58 | <L29Ah> | worksforme |
| 21:25:31 | <jakefromstatefar> | Hm, maybe it's nixes fault |
| 21:25:33 | <L29Ah> | try disabling stuff in User Interface setting |
| 21:25:46 | <L29Ah> | > Always On Top |
| 21:25:46 | <L29Ah> | looks sus |
| 21:25:47 | <lambdabot> | error: |
| 21:25:47 | <lambdabot> | Data constructor not in scope: Always :: t0 -> t1 -> terror: |
| 21:25:47 | <lambdabot> | • Data constructor not in scope: On |
| 21:29:14 | <jakefromstatefar> | Nix isn't entirely the problem, but having overlay enabled is wacky |
| 21:30:27 | <jakefromstatefar> | Changing workspaces still causes it to hide, but it returns when requested |
| 21:30:32 | <L29Ah> | ah, i never used the overlay |
| 21:30:49 | <L29Ah> | i think you should tell xmonad not to touch the overlay |
| 21:31:48 | <jakefromstatefar> | It doesn't even show up XD |
| 21:31:55 | <jakefromstatefar> | I'll just leave it disabled, I guess |
| 21:32:01 | <L29Ah> | btw https://github.com/l29ah/xmonad-config/blob/master/xmonad.hs#L79-L103 does this belong in xmonad-contrib? |
| 21:32:06 | <L29Ah> | liskin: ^ |
| 21:32:16 | <L29Ah> | pulls dbus |
| 21:33:10 | <geekosaur> | dbus probably makes it -extras |
| 21:35:31 | <jakefromstatefar> | Is there a way to make `XMonad.Layout.Hidden` work across workspaces? |
| 21:36:04 | <jakefromstatefar> | Currently, it only functions on a per-workspace basis. |
| 21:37:08 | <jakefromstatefar> | What I desire is: e.g - I hide/minimize window A on ws A, then try to unhide/unminimize on ws B, window A should appear. |
| 21:37:28 | <L29Ah> | https://hackage.haskell.org/package/xmonad-contrib-0.16/docs/XMonad-Layout-BoringWindows.html wtf is this |
| 21:37:40 | <L29Ah> | zero info in the docs on the actual impact on UI |
| 21:38:12 | <geekosaur> | L29Ah, it's best explained by the modules that use it, I think |
| 21:38:38 | <L29Ah> | that i need to grep, i guess :/ |
| 21:38:43 | <geekosaur> | jakefromstatefar, layout state only exists within a workspace |
| 21:39:14 | <geekosaur> | it would have to be redesigned to put the stack in XS instead of in the layout to work across workspaces like that |
| 21:39:57 | <geekosaur> | anyway a window marked boring can be auto-skipped by various move commands that are aware of the BoringWindows module |
| 21:40:14 | <geekosaur> | otherwise it has no effect |
| 21:41:35 | <jakefromstatefar> | boringWindows gives abilities to skip over windows marked as `boring` |
| 21:41:51 | <jakefromstatefar> | It doesn't work well with tabs |
| 21:42:00 | <jakefromstatefar> | Which's why I switched to `hidden` |
| 21:47:34 | <jakefromstatefar> | So, for 0.17, it looks like the only thing left is misc documentation/versioning/information updating, right? |
| 21:47:55 | <jakefromstatefar> | Are we going to merge some of my screenshots into xmonad-web? |
| 21:48:21 | <geekosaur> | no idea |
| 21:53:32 | <jakefromstatefar> | On another note: https://termbin.com/ogeb -- thoughts? |
| 21:53:50 | <jakefromstatefar> | It's a start to my version of the tree-structure proposal. |
| 21:56:18 | <liskin> | jakefromstatefar: "slotThe added this to the v0.17.1 milestone yesterday" at the bottom of https://github.com/xmonad/xmonad-contrib/issues/160 suggests that we aren't hoping to merge the screenshots before the 0.17.0 release, but would be nice to do it after |
| 21:56:37 | <liskin> | I must say that I'm sooo looking forward to a little holiday though :-) |
| 21:56:54 | <liskin> | (yeah I know it's not been that long since my last one, but it's been quite a crunch now) |
| 22:25:15 | × | seschwar quits (~seschwar@user/seschwar) (Quit: :wq) |
| 22:27:14 | <liskin> | Solid: regarding the X.L.Fullscreen warning thing, we might just ignore it and fix it only if people start complaining about it; perhaps some of them will be able to migrate away from using a build script |
| 22:27:30 | <liskin> | Solid: but there is a risk that there will be lots of them :-D |
| 22:28:54 | <liskin> | or I can just duplicate the addSupported function, that's not a big deal |
| 23:10:06 | × | slep quits (~slep@cpc150002-brnt4-2-0-cust437.4-2.cable.virginm.net) (Ping timeout: 245 seconds) |
| 23:47:28 | → | mvk joins (~mvk@2607:fea8:5cc1:300::7f76) |
| 23:56:01 | spider_ | is now known as spider |
| 23:58:46 | → | a6a45081-2b83 joins (~aditya@2601:249:4300:1296:88ec:cc73:84d4:1507) |
All times are in UTC on 2021-10-24.