Logs: freenode/#xmonad
| 2021-02-08 15:55:57 | × | xaltsc quits (~xaltsc@unaffiliated/xaltsc) (Ping timeout: 260 seconds) |
| 2021-02-08 15:56:35 | <mc47> | now it decided to work even without setting the env variable.. fine by me |
| 2021-02-08 15:59:13 | × | tremon quits (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) (Quit: getting boxed in) |
| 2021-02-08 15:59:29 | → | tremon joins (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) |
| 2021-02-08 16:00:00 | × | tremon quits (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) (Client Quit) |
| 2021-02-08 16:00:14 | → | tremon joins (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) |
| 2021-02-08 16:03:11 | <Liskni_si> | You sure it's not set? |
| 2021-02-08 16:04:11 | × | evanjs quits (~evanjs@075-129-098-007.res.spectrum.com) (Quit: ZNC 1.8.2 - https://znc.in) |
| 2021-02-08 16:04:58 | → | evanjs joins (~evanjs@075-129-098-007.res.spectrum.com) |
| 2021-02-08 16:07:46 | <mc47> | I was wrong, it still acts weird when it's not set, and it works when it's set |
| 2021-02-08 16:08:07 | <mc47> | it was tiled so I didn't see the big grey area that was unusable |
| 2021-02-08 16:13:54 | × | evanjs quits (~evanjs@075-129-098-007.res.spectrum.com) (Quit: ZNC 1.8.2 - https://znc.in) |
| 2021-02-08 16:15:05 | × | tremon quits (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) (Quit: getting boxed in) |
| 2021-02-08 16:15:38 | → | tremon joins (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) |
| 2021-02-08 16:15:41 | → | evanjs joins (~evanjs@075-129-098-007.res.spectrum.com) |
| 2021-02-08 16:16:21 | × | tremon quits (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) (Client Quit) |
| 2021-02-08 16:16:49 | <Solid> | oh-java.png |
| 2021-02-08 16:17:01 | → | tremon joins (~aschuring@217-63-61-89.cable.dynamic.v4.ziggo.nl) |
| 2021-02-08 16:19:32 | <Solid> | I'd be wary of linking to the xmonad wiki in docs at this point, seeing that we may soon lose it (at least in its current form) |
| 2021-02-08 16:25:42 | <mc47> | That slipped my mind |
| 2021-02-08 16:52:54 | × | joshproehl quits (~quassel@dtools-d10-tx-1.inf.daedalusdreams.com) (Quit: Disappearing act!) |
| 2021-02-08 16:53:13 | → | joshproehl joins (~quassel@dtools-d10-tx-1.inf.daedalusdreams.com) |
| 2021-02-08 16:57:18 | → | Nahra joins (~Nahra@unaffiliated/nahra) |
| 2021-02-08 16:58:19 | × | joshproehl quits (~quassel@dtools-d10-tx-1.inf.daedalusdreams.com) (Quit: Disappearing act!) |
| 2021-02-08 16:58:36 | → | joshproehl joins (~quassel@dtools-d10-tx-1.inf.daedalusdreams.com) |
| 2021-02-08 16:58:59 | → | growpotkin joins (~growpotki@130-45-30-154.dyn.grandenetworks.net) |
| 2021-02-08 17:00:46 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-02-08 17:01:03 | <geekosaur> | so far, so good… |
| 2021-02-08 17:04:10 | <geekosaur> | mc47, the LG3D hack is specific to Oracle Java, the _JAVA_AWT_VM_NONREPARENTING is specific to openjdk, other implementations may have their own ways |
| 2021-02-08 17:05:27 | <Liskni_si> | I'd expect even Oracle java to support _JAVA_AWT_VM_NONREPARENTING by now. (But I'm not going to check that.) |
| 2021-02-08 17:05:37 | <geekosaur> | I wouldn't, because Oracle |
| 2021-02-08 17:05:59 | <geekosaur> | not like they support awesome or xmonad |
| 2021-02-08 17:07:25 | <geekosaur> | anyway I did a clean startup and didn't have to wait for it to rebuild anything or even the odd slowness of the first time it made the exe link, so I guess it's working and I can document both my versions |
| 2021-02-08 17:07:25 | <Liskni_si> | it'd be incredibly silly to diverge from openjdk so much |
| 2021-02-08 17:08:11 | <geekosaur> | you seriously think Oracle thinks that way? their way of thinking is that openjdk should not diverge from Them so much |
| 2021-02-08 17:11:07 | <geekosaur> | (openjdk is not the inheritor of Sun's IP rights in Java; Oracle bought the remains of Sun specifically for the Java IP) |
| 2021-02-08 17:13:31 | <mc47> | geekosaur, do you think the LG3D hack should also be added? |
| 2021-02-08 17:14:32 | <mc47> | Why do things get really confusing really fast? It's so weird that this bug exists in the first place |
| 2021-02-08 17:14:32 | <geekosaur> | that one's pretty much legendary (and about as relevant to the modern world :) — I suspect anyone who cares is no longer using a reparenting wm at all |
| 2021-02-08 17:14:57 | <geekosaur> | oh, I know why the bug exists (and neither hack works with mono which has the same problem) |
| 2021-02-08 17:15:23 | <mc47> | I'm intrested, why? |
| 2021-02-08 17:15:29 | <geekosaur> | the problem is cross-platform window toolkits, because Windows uses screen-relative positioning but X11 is window-relative |
| 2021-02-08 17:16:21 | <geekosaur> | most cross-platform toolkits emulate Windows, so they have to translate X11 window-relative positions to screen-relative. but this means traversing back to the root window, which needs an extra step if the window has been reparented |
| 2021-02-08 17:16:33 | <geekosaur> | and if that step is present it gets confused if windows are *not* reparented |
| 2021-02-08 17:17:25 | <geekosaur> | (mono definitely emulates windows since it's .net emulation for linux/unix) |
| 2021-02-08 17:18:19 | <mc47> | hmm, I see |
| 2021-02-08 17:18:27 | <mc47> | thanks! |
| 2021-02-08 17:32:37 | <Solid> | I'm starting a log with "wise geekosaur explanations" |
| 2021-02-08 17:33:30 | <Solid> | good first entry I'd say :> |
| 2021-02-08 17:47:21 | × | ChubaDuba quits (~ChubaDuba@37.112.228.11) (Quit: WeeChat 1.6) |
| 2021-02-08 18:02:42 | × | geekosaur quits (82650c7a@130.101.12.122) (Ping timeout: 240 seconds) |
| 2021-02-08 18:17:45 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-02-08 18:20:53 | <Liskni_si> | makes me wonder why there's no XMonad.Hook.Reparent yet :-) |
| 2021-02-08 18:21:55 | <geekosaur> | right now there are things intermixed that would want the frame window vs. the client window, that would need to be untangled for reparenting to work |
| 2021-02-08 18:22:22 | <geekosaur> | it'd be a fairly big change to core and as big or bigger for some contribs (Decoration comes to mind) |
| 2021-02-08 18:22:59 | <geekosaur> | Decoration might be especially bad because most uses would want to use the frame window… but then there's how Tabbed uses it |
| 2021-02-08 18:24:29 | <Liskni_si> | oh if it can't be done in a hacky way then nevermind :-0) |
| 2021-02-08 18:24:35 | <Liskni_si> | s/0// |
| 2021-02-08 18:35:29 | <Solid> | I guess no one who uses xmonad is invested enough in java to go through the pain :) |
| 2021-02-08 18:37:37 | <Liskni_si> | I remember that a couple years ago I had a LD_PRELOAD wrapper to fix josm (which uses AWT I think) not working properly with xinerama |
| 2021-02-08 18:37:59 | <Liskni_si> | so I was somewhat invested, but didn't really need to investigate the reparenting route |
| 2021-02-08 18:58:25 | → | xaltsc joins (~xaltsc@unaffiliated/xaltsc) |
| 2021-02-08 19:12:31 | × | davemq quits (~davemq@99-179-0-50.lightspeed.austtx.sbcglobal.net) (Read error: Connection reset by peer) |
| 2021-02-08 19:14:33 | → | davemq joins (~davemq@2600:1700:b1c0:2580::4d8) |
| 2021-02-08 19:15:42 | × | geekosaur quits (82650c7a@130.101.12.122) (Ping timeout: 240 seconds) |
| 2021-02-08 19:20:30 | × | Nahra quits (~Nahra@unaffiliated/nahra) (Quit: leaving) |
| 2021-02-08 19:21:59 | <mc47> | How would someone debug the startupHook? Like how can I log to xmonad.errors for example? |
| 2021-02-08 19:26:53 | × | lambdabot quits (~lambdabot@haskell/bot/lambdabot) (Remote host closed the connection) |
| 2021-02-08 19:28:46 | <Liskni_si> | if you invoke xmonad via startx or xinit, then its stdout/stderr is in ~/.xsession-errors and you can just "print" or "io $ print …" |
| 2021-02-08 19:29:14 | → | lambdabot joins (~lambdabot@haskell/bot/lambdabot) |
| 2021-02-08 19:29:14 | ChanServ | sets mode +v lambdabot |
| 2021-02-08 19:29:30 | <Liskni_si> | or you can add some systemd-cat or systemd-run somewhere and have that output in the journal with additional metadata like time and pid |
| 2021-02-08 19:30:54 | × | hexo quits (~hexo@gateway/tor-sasl/hexo) (Remote host closed the connection) |
| 2021-02-08 19:31:12 | → | hexo joins (~hexo@gateway/tor-sasl/hexo) |
| 2021-02-08 19:33:41 | → | geekosaur joins (82650c7a@130.101.12.122) |
| 2021-02-08 19:33:42 | <mc47> | I'm letting KDE worry about invoking xmonad |
| 2021-02-08 19:36:08 | <geekosaur> | I've written a bunch of debug hooks but they all assume stderr is going somewhere useful. where that is will depend |
| 2021-02-08 19:36:43 | <geekosaur> | on my systems (all debian-derived) it's .xsession-errors. on fedora it's supposedly under control of journalctl, iirc |
| 2021-02-08 19:37:09 | <mc47> | I have no idea where that goes |
| 2021-02-08 19:37:27 | <geekosaur> | in other news, I found https://wiki.haskell.org/Special:Export |
| 2021-02-08 19:37:40 | <mc47> | I could just run strace for write syscalls and grep what I want |
| 2021-02-08 19:37:46 | <geekosaur> | possibly pandoc can do something useful with the result |
| 2021-02-08 19:38:07 | <geekosaur> | strace is possibly the most painful way of doing this, especially since there'll be 100 signals per second from the runtime |
| 2021-02-08 19:38:54 | <geekosaur> | also lots of writes to the X server socket |
| 2021-02-08 19:38:59 | <Solid> | Liskni_si: afaik this is not standardized |
| 2021-02-08 19:39:30 | <Solid> | my system sends the info god knows where by default and I have to explicitly redirect it so some errors file via something like `exec launchx >& ~/.xsession.log' |
| 2021-02-08 19:39:42 | <mc47> | it is painful, but it might just work |
| 2021-02-08 19:40:12 | <mc47> | unless someone has an idea on what KDE does with window managers? |
| 2021-02-08 19:40:23 | <geekosaur> | it may be possible to make strace itself limit to writes to stderr |
| 2021-02-08 19:40:31 | <geekosaur> | no idea with recent kde, sorry |
| 2021-02-08 19:40:43 | <geekosaur> | last time I played with it was kde2 :) |
| 2021-02-08 19:40:56 | <geekosaur> | things have… changed since then |
| 2021-02-08 19:41:29 | <Solid> | surely kde must have this documented somewhere? |
| 2021-02-08 19:43:13 | <geekosaur> | I was going to say last time I needed to know someting about kde I had to dig in the source, but now that I think about it that was actually something about xcb :/ |
| 2021-02-08 19:50:09 | <Liskni_si> | Solid: oh, right, it's debian's Xsession that's doing the redirect here |
| 2021-02-08 19:51:34 | <geekosaur> | doesn't debian use .xsession-errors though? ubuntu and mint do but it may be inherited from ubuntu instead of from debian |
| 2021-02-08 19:52:48 | <mc47> | I also have no idea, and I'm not even sure I could replicate my current setup again |
| 2021-02-08 19:53:00 | <mc47> | After a lot of trying, it just "works" |
| 2021-02-08 19:53:27 | geekosaur | at least tries to keep notes |
| 2021-02-08 19:53:58 | <geekosaur> | but I missed one this time around and had to try to reconstruct from the wiki, where I again missed one (xmonad vs. start-xmonad) |
All times are in UTC.