| 17 May 2025 |
maralorn | At least the header indicates that. | 12:44:06 |
maralorn | I wonder if it works if you just rm it in the postPatch phase. | 12:44:33 |
ners | It does! Thanks for the tip. :) | 12:46:37 |
ners | I'll open a PR against haskell-updates. | 12:47:01 |
ners | * I'll open a PR against haskell-updates to unbreak it. | 12:47:06 |
maralorn | I guess this has something to do with the recent modification of cabal build types, yet the error message is a bit baffling. | 12:47:38 |
ners | Do you think it makes sense to open a PR upstream removing the Setup.hs? I never quite understood what purpose that file serves. | 12:48:13 |
winston | my understanding is that it's basically the equivalent of a Makefile? like running
runhaskell Setup configure
runhaskell Setup build
runhaskell Setup install
| 12:51:29 |
winston | i actually ran into issues not having a Setup.hs present when I tried to package some of my code for other Distros like Arch | 12:51:48 |
maralorn | I feel semi comfortable giving advice on this without understanding the problem. 😄 But yeah my general heuristic is that packages which don’t need special setup shouldn’t ship their own Setup.hs. The nixpkgs builder uses a fallback when that happens. | 12:54:25 |
maralorn | But it would be interesting to see how their Setup.hs deviates from the current default. | 12:55:19 |
ners | https://github.com/facebook/Haxl/blob/main/Setup.hs
It has a licence header on top. :think | 13:29:01 |
ners | * https://github.com/facebook/Haxl/blob/main/Setup.hs
It has a licence header on top. 🤔 | 13:29:07 |
maralorn | Plot twist: Someone somewhere built in a killswitch triggered by Meta. 😆 | 13:34:18 |
alexfmpe | probably company-provisioned editor config | 20:47:28 |
alexfmpe | big tech has really retentive lawyers | 20:48:14 |
alexfmpe | facebook/duckling wouldn't even accept my PR fixing the syntax for a haddock comment without me signing their snowflake license | 20:48:31 |
alexfmpe | proportional much | 20:49:04 |
| NixOS Moderation Bot banned @mightyiam:matrix.org (persistent time-wasting). | 23:30:21 |
| 18 May 2025 |
wildwestrom | Thank you for the response, I'm currently trying this out. Each time I build it takes about 20 minutes, so I'll let you know how it goes in awhile. | 11:22:35 |
wildwestrom | After an update, I've come up against another failure.
❯ nix build
warning: Git tree '/home/main/code/community/lamdu' is dirty
error:
… while calling the 'derivationStrict' builtin
at <nix/derivation-internal.nix>:37:12:
36|
37| strict = derivationStrict drvAttrs;
| ^
38|
… while evaluating derivation 'Lamdu-0.8.1'
whose name attribute is located at /nix/store/g3z13ffr1wzw8m9wivqyni2nbjl956wj-source/pkgs/stdenv/generic/make-derivation.nix:480:13
… while evaluating attribute 'buildInputs' of derivation 'Lamdu-0.8.1'
at /nix/store/g3z13ffr1wzw8m9wivqyni2nbjl956wj-source/pkgs/stdenv/generic/make-derivation.nix:535:13:
534| depsHostHost = elemAt (elemAt dependencies 1) 0;
535| buildInputs = elemAt (elemAt dependencies 1) 1;
| ^
536| depsTargetTarget = elemAt (elemAt dependencies 2) 0;
(stack trace truncated; use '--show-trace' to show the full, detailed trace)
error: function 'anonymous lambda' called with unexpected argument 'libxml'
at /nix/store/zjbqzwnsp7092jagr1xgrd33dq1i35xp-cabal2nix-test-framework-0.8.2.2/default.nix:1:1:
1| { mkDerivation, ansi-terminal, ansi-wl-pprint, base, bytestring
| ^
2| , containers, hostname, HUnit, lib, old-locale, QuickCheck, random
| 11:38:59 |
wildwestrom | Here's the full trace. https://bpa.st/QC7A | 11:40:43 |
sterni (he/him) | looks like the hard coded overrides are outdated in stacklock2nix https://github.com/cdepillabout/stacklock2nix/blob/108f6ab6d0208e3842443bea26dfd72b37d82f28/nix/build-support/stacklock2nix/cabal2nixArgsForPkg.nix#L128 | 13:23:56 |
wildwestrom | Damn, alright. Guess I'll need a different approach. | 13:24:38 |
sterni (he/him) | As is documented in that file you can pass your own overrides. | 13:25:42 |
wildwestrom | overrides = final.lib.composeManyExtensions [
(oldAttrs.overrides or (_: _: { }))
final.lamdu-haskell-stacklock.stackYamlResolverOverlay
final.lamdu-haskell-stacklock.stackYamlExtraDepsOverlay
final.lamdu-haskell-stacklock.stackYamlLocalPkgsOverlay
final.lamdu-haskell-stacklock.suggestedOverlay
# (hfinal: hprev: {
# language-ecmascript = final.haskell.lib.dontCheck hprev.language-ecmascript;
# text-trie = final.haskell.lib.dontCheck hprev.text-trie;
# rocksdb-haskell = final.haskell.lib.dontCheck hprev.rocksdb-haskell;
# bindings-GLFW = final.haskell.lib.dontCheck hprev.bindings-GLFW;
# GLFW-b = final.haskell.lib.dontCheck hprev.GLFW-b;
# })
];
I assume somewhere in here yeah?
This is my first time doing this. What would I put?
| 13:33:55 |
sterni (he/him) | it's documented here https://github.com/cdepillabout/stacklock2nix/blob/108f6ab6d0208e3842443bea26dfd72b37d82f28/nix/build-support/stacklock2nix/default.nix#L27-L31 | 13:41:21 |
wildwestrom | Ok, that got me through the first couple errors. Now there's an infinite recursion. Not sure what to do about that.
Here's what I'm working with: https://github.com/lamdu/lamdu/pull/241
Full Trace: https://bpa.st/GK4A | 14:42:39 |
wildwestrom | Goddamn, I fix one thing and another thing gets broken. | 16:50:03 |
wildwestrom | I think I got it! | 21:06:11 |