| 11 Nov 2025 |
 ElvishJerricco | probably not but I'm afraid to change it :P | 11:25:08 |
 Grimmauld (migrated to @grimmauld:m.grimmauld.de) | like, that wrapper does nothing? i think? | 11:25:29 |
| ElvishJerricco | yea wtf | 11:25:44 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | it removes pkgs, thats about it | 11:25:52 |
| ElvishJerricco | I mean | 11:26:09 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | it inherits name and paths, it explicitly does extra outputs | 11:26:11 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | so why do we need that at all | 11:26:24 |
| ElvishJerricco | technically, the args given to this buildEnv' are more powerful, because they become derivation attrs, not args to buildEnv | 11:26:26 |
| ElvishJerricco | so that's technically a difference | 11:26:30 |
| ElvishJerricco | but I don't know if that's actually used? | 11:26:40 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | HUH | 11:26:49 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | okay that is actually fair, hold on let me check | 11:27:02 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | oh yeah there is stuff like postBuild, meta, outputs and passthru and a whole mess of other things:
https://github.com/NixOS/nixpkgs/blob/addd85b5d06ea90d5a7923956a65baf96bcd1e9f/pkgs/tools/typesetting/tex/texlive/build-tex-env.nix#L414-L480 | 11:28:08 |
| ElvishJerricco | ok, so unless anyone wants to go sifting through all those instances and finding a better way to do them, we should just keep the override | 11:28:45 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) |
# no indent for git diff purposes
treewide nixfmt says hello
| 11:28:54 |
| ElvishJerricco | so the thing vcunat and I did is probably the best thing to do for now | 11:28:58 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | i mean there is only 3 instances | 11:30:08 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | its in a let...in block and not reexported | 11:30:23 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | but yeah, do the workaround | 11:30:54 |
| ElvishJerricco | yea I mean it looks like we'd just end up doing the overrideAttrs in those 3 sites instead, so there's no real benefit to trying to get rid of it | 11:31:29 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | or alternatively hand-roll a pseudo-buildenv using stdenv.mkDerivation that has the power to do these things | 11:32:22 |
| Grimmauld (migrated to @grimmauld:m.grimmauld.de) | not sure either is better | 11:32:42 |
| ElvishJerricco | that seems like more code / effort for no benefit | 11:33:42 |
| ElvishJerricco | vcunat: you want to author the commit / PR or do you want me to? | 11:35:02 |
 Vladimír Čunát | I don't care really. | 11:35:24 |
| Vladimír Čunát | But I'm relatively busy today. | 11:35:30 |
| Vladimír Čunát | (with work stuff) | 11:35:36 |
| Vladimír Čunát | So either way I won't have capacity for longer discussion about the changes. | 11:36:53 |
| ElvishJerricco | https://github.com/NixOS/nixpkgs/pull/460639 | 11:39:43 |
 winston | I was looking into the arrow-cpp test failure on darwin, strangely the binary of the tests that fail end up with two versions of protobuf in the dynamic linker paths
v31, which is what we want in the derivation, v33 coming from somewhere else
whats stranger is that master also shows v32, but it doesn't have the same runtime error
I compiled it with -fsanitize=address and that pointed me into this direction: https://github.com/protocolbuffers/protobuf/issues/5290
https://gist.github.com/nekowinston/f6cbea976ba4605f318b9329d96a0544 | 11:52:38 |
