2025/04/04

Newest at the top

2025-04-04 22:05:55 +0200peterbecich(~Thunderbi@syn-047-229-123-186.res.spectrum.com) peterbecich
2025-04-04 22:02:20 +0200Eoco(~ian@128.101.131.218) Eoco
2025-04-04 22:01:59 +0200remexre(~remexre@user/remexre) remexre
2025-04-04 22:00:03 +0200Eoco(~ian@128.101.131.218) (Ping timeout: 268 seconds)
2025-04-04 21:58:43 +0200remexre(~remexre@user/remexre) (Ping timeout: 245 seconds)
2025-04-04 21:38:34 +0200Adran(~adran@botters/adran) Adran
2025-04-04 21:36:28 +0200 <merijn> Actually...maybe I can get away just commenting out the paths and relying on the defaults?
2025-04-04 21:35:25 +0200 <merijn> I mean, let's face it, 98% of the use case is covered by supporting $HOME and/or ~
2025-04-04 21:34:34 +0200 <sclv> i agree with you that coupling "let people use some env variables in cabal config" to "xdg implementation" made things more intertwingled
2025-04-04 21:33:52 +0200 <sclv> commented on the other ticket, though i don't have much time for cabal these days. with the xdg stuff cleared up i think just having some set of env variables that are expanded (and also kept in our rebuild-checking logic) should be straightforward.
2025-04-04 21:32:18 +0200 <merijn> But it got stuck in some bikeshedding hell and I never got consensus on how to do it
2025-04-04 21:32:00 +0200 <merijn> sclv: That's why I had it tracked :p
2025-04-04 21:31:16 +0200 <sclv> merijn: funny to see that the issue came from a discussion with you and was assigned to you even then :-/
2025-04-04 21:31:14 +0200rit(~rit@2409:40e0:39:317c:5ce4:875:e0a7:d628) (Remote host closed the connection)
2025-04-04 21:31:09 +0200peterbecich(~Thunderbi@syn-047-229-123-186.res.spectrum.com) (Ping timeout: 252 seconds)
2025-04-04 21:30:23 +0200jacopovalanzano(~jacopoval@cpc151911-cove17-2-0-cust105.3-1.cable.virginm.net) (Quit: Client closed)
2025-04-04 21:29:50 +0200 <merijn> I mean, I guess it's still a duplicate of hvr's issue, but for a second there I had hope
2025-04-04 21:29:17 +0200 <sclv> yeah, i think that was a mistake, agreed
2025-04-04 21:28:17 +0200 <merijn> I was hoping that in 2025 I could *finally* have a cabal config shared across my machines >.>
2025-04-04 21:27:30 +0200 <merijn> It was closed because of Athas' XDG implementation, but it's IMO entirely unrelated to XDG
2025-04-04 21:27:02 +0200 <merijn> sclv: https://github.com/haskell/cabal/issues/5998
2025-04-04 21:25:43 +0200 <sclv> can you link to the issue? but yeah, i don't think i've ever seen it done
2025-04-04 21:25:29 +0200 <merijn> hmm, the more I look into it, it's just a lie that my issue was closed as fixed
2025-04-04 21:22:22 +0200 <merijn> My issue to support that was closed as implemented, but I can't actually find any documentation on it
2025-04-04 21:22:06 +0200 <merijn> Does anyone know how to use environment variables like $HOME in cabal's config?
2025-04-04 21:21:56 +0200wlhn(~wlhn@2001-14ba-4e04-ccf0-9803-c364-b8d-97b4.rev.dnainternet.fi) (Quit: Leaving)
2025-04-04 21:21:34 +0200 <merijn> \o/
2025-04-04 21:21:25 +0200machinedgod(~machinedg@d108-173-18-100.abhsia.telus.net) machinedgod
2025-04-04 21:18:12 +0200 <kqr> merijn, Oh, that was very helpful. It also made clear what happened: laziness caused some of the work from the first action to be performed during the second action evaluation. I should have known!
2025-04-04 21:13:09 +0200kimiamania(~65804703@user/kimiamania) kimiamania
2025-04-04 21:12:43 +0200tavare(~tavare@user/tavare) (Remote host closed the connection)
2025-04-04 21:11:37 +0200kimiamania(~65804703@user/kimiamania) (Quit: PegeLinux)
2025-04-04 21:10:54 +0200tavare(~tavare@user/tavare) tavare
2025-04-04 21:10:54 +0200tavare(~tavare@150.129.88.189) (Changing host)
2025-04-04 21:10:54 +0200tavare(~tavare@150.129.88.189) tavare
2025-04-04 21:10:44 +0200Adran(~adran@botters/adran) (Ping timeout: 272 seconds)
2025-04-04 21:09:23 +0200peterbecich(~Thunderbi@syn-047-229-123-186.res.spectrum.com) peterbecich
2025-04-04 21:07:33 +0200 <merijn> Entirely unrelatedly: In the absence of a tree-sitter grammar, does anyone have/know a decent syntax file for .cabal?
2025-04-04 21:06:38 +0200 <merijn> kqr: https://hackage.haskell.org/package/base-4.21.0.0/docs/GHC-Exts.html#v:traceMarker-35-
2025-04-04 21:05:38 +0200 <merijn> kqr: Can also do manual cost centre annotation
2025-04-04 21:05:10 +0200 <merijn> kqr: You can have GHC inject labels and instruct the profiler to use those markers
2025-04-04 21:04:12 +0200 <kqr> I have a program that performs two actions. One of them is known to be very expensive, and then there's another that seems to be slower than I thought it would be. I want to profile only the latter, but with regular RTS profiling the former ovbeing slow makes it hard to tease out which parts of the latter is slow. What do I do?
2025-04-04 21:00:47 +0200caconym(~caconym@user/caconym) caconym
2025-04-04 21:00:02 +0200caconym(~caconym@user/caconym) (Quit: bye)
2025-04-04 20:56:10 +0200lxsameer(~lxsameer@Serene/lxsameer) (Ping timeout: 272 seconds)
2025-04-04 20:47:36 +0200acidjnk_new3(~acidjnk@p200300d6e71c4f05ad6cd39b235d43c6.dip0.t-ipconnect.de)
2025-04-04 20:47:05 +0200Maxdamantus(~Maxdamant@user/maxdamantus) Maxdamantus
2025-04-04 20:46:49 +0200Maxdamantus(~Maxdamant@user/maxdamantus) (Ping timeout: 260 seconds)
2025-04-04 20:40:45 +0200merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn
2025-04-04 20:40:09 +0200merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 244 seconds)