2025/08/15

Newest at the top

2025-08-15 23:09:17 +0200L29Ah(~L29Ah@wikipedia/L29Ah) L29Ah
2025-08-15 22:39:23 +0200L29Ah(~L29Ah@wikipedia/L29Ah) (Read error: Connection reset by peer)
2025-08-15 22:36:31 +0200haskellbridge(~hackager@syn-096-028-224-214.res.spectrum.com) hackager
2025-08-15 22:34:12 +0200haskellbridge(~hackager@syn-096-028-224-214.res.spectrum.com) (Remote host closed the connection)
2025-08-15 20:05:12 +0200werneta(~werneta@syn-071-083-160-242.res.spectrum.com) werneta
2025-08-15 18:05:43 +0200Digit(~user@user/digit) Digit
2025-08-15 18:04:39 +0200Digit(~user@user/digit) (Read error: Connection reset by peer)
2025-08-15 15:24:44 +0200tv(~tv@user/tv) tv
2025-08-15 14:34:26 +0200 <liskin> Well there's so many things I should do
2025-08-15 14:34:12 +0200 <liskin> Hm, the "live" docs probably use old ghc. I should fix that.
2025-08-15 13:31:26 +0200ft(~ft@p508db692.dip0.t-ipconnect.de) ft
2025-08-15 12:29:36 +0200L29Ah(~L29Ah@wikipedia/L29Ah) L29Ah
2025-08-15 10:39:27 +0200L29Ah(~L29Ah@wikipedia/L29Ah) (Ping timeout: 252 seconds)
2025-08-15 08:21:19 +0200ft(~ft@p508db692.dip0.t-ipconnect.de) (Quit: leaving)
2025-08-15 06:17:53 +0200haskellbridge(~hackager@syn-096-028-224-214.res.spectrum.com) hackager
2025-08-15 06:17:10 +0200haskellbridge(~hackager@syn-096-028-224-214.res.spectrum.com) (Remote host closed the connection)
2025-08-15 05:53:29 +0200 <haskellbridge> <iqubic (she/her)> I know Haskell well enough to understand the XMonad code, but actually learning all the library functions and types involved to make this work will take me a long while. I've figured out where NixOS put its session output, so I can use this module as it currently is.
2025-08-15 05:52:07 +0200 <geekosaur> I've never been very happy about people having to figure out where their session output is (if it's there at all: some systems cap it via cgroups) to use it
2025-08-15 05:51:17 +0200 <geekosaur> I want to
2025-08-15 05:51:04 +0200 <haskellbridge> <iqubic (she/her)> Anyways, what exists now works for me. You don't need to worry about spawning a new window for ManageDebug output unless you want to.
2025-08-15 05:48:42 +0200 <geekosaur> which is why the type has a MonadIO constraint
2025-08-15 05:48:34 +0200 <geekosaur> "io" is short for liftIO
2025-08-15 05:47:17 +0200 <haskellbridge> <iqubic (she/her)> Wait... is that still running unsafePerformIO? I can't tell.
2025-08-15 05:46:36 +0200 <haskellbridge> <iqubic (she/her)> https://xmonad.github.io/xmonad-docs/xmonad/src/XMonad.Core.html#trace
2025-08-15 05:46:07 +0200 <geekosaur> I guess that means I can get away with plumbing a Handle to an xmessage through there after all
2025-08-15 05:45:23 +0200 <geekosaur> hah. it's from XMonad.Core
2025-08-15 05:44:52 +0200 <haskellbridge> <iqubic (she/her)> I'm going to assume this is some weirdness related to X.C.Prime and just not worry about this.
2025-08-15 05:44:16 +0200 <geekosaur> yeh, I just saw that
2025-08-15 05:44:08 +0200 <haskellbridge> <iqubic (she/her)> Hovering my mouse over "trace" there shows me "MonadIO m => m ()" as the signature.
2025-08-15 05:43:28 +0200 <haskellbridge> <iqubic (she/her)> https://xmonad.github.io/xmonad-docs/xmonad-contrib/src/XMonad.Hooks.ManageDebug.html#debugManageH…
2025-08-15 05:42:50 +0200 <geekosaur> that should be it, maybe we do have a different trace in scope
2025-08-15 05:41:53 +0200 <haskellbridge> <iqubic (she/her)> But I found this: https://xmonad.github.io/xmonad-docs/xmonad-contrib/
2025-08-15 05:41:48 +0200 <haskellbridge> <iqubic (she/her)> Not sure if this is the live docs, or just a different domain showing the same haddocks.
2025-08-15 05:40:25 +0200 <geekosaur> not currently
2025-08-15 05:38:37 +0200 <haskellbridge> <iqubic (she/her)> Is there a new release planned?
2025-08-15 05:38:25 +0200 <geekosaur> I can never remember where they live
2025-08-15 05:38:16 +0200 <geekosaur> I think if you use (and can find) the live docs it's been fixed and will go live with the next release
2025-08-15 05:37:37 +0200 <geekosaur> but haddock insists on linking to everything it exports
2025-08-15 05:37:21 +0200 <geekosaur> that's a bug in how we have things set up. X.C.Prime is an ancient experiment that went nowhere
2025-08-15 05:36:04 +0200 <haskellbridge> <iqubic (she/her)> I think that's not the same "trace" as DebugManage uses.
2025-08-15 05:35:49 +0200 <haskellbridge> <iqubic (she/her)> I was viewing the source code on the Hackage site and the mouse hover info was showing me the signature for this "trace" https://hackage.haskell.org/package/xmonad-contrib-0.18.1/docs/XMonad-Config-Prime.html#v:trace
2025-08-15 05:32:49 +0200 <haskellbridge> <iqubic (she/her)> Oh.. So that's not actually "real" I/O. Got it. I see now why opening a window for debug output would be hard.
2025-08-15 05:32:16 +0200 <geekosaur> which is why it has to use debugBelch: you run into problems with locked Handles otherwise
2025-08-15 05:31:58 +0200 <geekosaur> trace uses unsafePerformIO
2025-08-15 05:30:45 +0200 <haskellbridge> <iqubic (she/her)> Or is that a different because "trace" is "MonadIO" action and not an "X Monad" action?
2025-08-15 05:30:31 +0200 <dpn> *grabs popcorn*
2025-08-15 05:30:03 +0200 <haskellbridge> <iqubic (she/her)> geekosaur: Isn't calling "trace" technically doing I/O?
2025-08-15 05:29:08 +0200 <dpn> being able to summon a window for the log would be pretty amazing
2025-08-15 05:23:33 +0200 <geekosaur> I've been thinking about how to capture output to a window without having to plumb I/O through the pure parts of a manageHook
2025-08-15 05:23:12 +0200 <haskellbridge> <iqubic (she/her)> Ah. I se.