!kjdutkOsheZdjqYmqp:nixos.org

Nixpkgs / NixOS contributions

1890 Members
NixOS 24.05 Uakari | #review-requests:nixos.org | https://nixos.org/blog/announcements.html#nixos-23.11 | https://hydra.nixos.org/jobset/nixos/trunk-combined | https://reproducible.nixos.org/ | 24.05 RMs: wegank & Mic92418 Servers

Load older messages

SenderMessageTime
22 Oct 2024
@lily:lily.flowersLily Foster
In reply to @aftix:matrix.org
is there any sort of guidelines for packaging things that don't have version tags / releases in git? specifically, some rust programs which don't have any sort of releases but usually have a version in Cargo.toml
see https://github.com/NixOS/nixpkgs/blob/master/pkgs/README.md#versioning for versioning guidelines. that there is a version number in Cargo.toml probably does not change the guidelines for packages that don't have tagged releases there, but more concrete context may help for suggesting what to do 		23:03:52
@adam:robins.wtfadamcstephens joined the room.23:20:10
@adam:robins.wtfadamcstephens šŸ‘‹ 23:25:52
23 Oct 2024
@niko:nyanbinary.rsnyanbinary šŸ³ļøā€āš§ļø left the room.01:15:44
@niko:nyanbinary.rsnyanbinary šŸ³ļøā€āš§ļø joined the room.01:17:45
@julius:mtx.liftm.deJCaesar Btw, what's the policy for marking a package as broken on master? guickgui has been broken for 8 weeks, there's no maintainer reaction to the related issue for 4, and it keeps annoying me by popping up in my nixpkgs-reviewsā€¦ Is that enough? 01:44:58
@julius:mtx.liftm.deJCaesar

happysalada: Thanks for bumping vector.

Btw, one thing I noticed while looking at the package def is that even when enableKafka is disabled, the features setup of vector still enables the rdkafka feature (e.g. transitively through sinks or sources). So to circument the ? issue, you could just add rdkafka/gssapi-vendored to the feature list and get rid of that confusing parameter, I think. (Personally, I'd remove the parameters entirely, the feature set is insanely complex and whoever wants to touch it anyway can do so through .overrideAttrs.)

02:35:08
@julius:mtx.liftm.deJCaesar *

happysalada: Thanks for bumping+fixing vector.

Btw, one thing I noticed while looking at the package def is that even when enableKafka is disabled, the features setup of vector still enables the rdkafka feature (e.g. transitively through sinks or sources). So to circument the ? issue, you could just add rdkafka/gssapi-vendored to the feature list and get rid of that confusing parameter, I think. (Personally, I'd remove the parameters entirely, the feature set is insanely complex and whoever wants to touch it anyway can do so through .overrideAttrs.)

02:35:32
@julius:mtx.liftm.deJCaesar *

happysalada: Thanks for bumping+fixing vector.

Btw, one thing I noticed while looking at the package def is that even when enableKafka is disabled, the features setup of vector still enables the rdkafka feature (e.g. transitively through sinks or sources). So to circument the ? issue, you could just add rdkafka/gssapi-vendored to the default feature list and get rid of that confusing parameter, I think. (Personally, I'd remove the parameters entirely, the feature set is insanely complex and whoever wants to touch it anyway can do so through .overrideAttrs.)

02:39:58
@aftix:matrix.orgaftixanyone else using jj with nixpkgs? I can't get my jj-git colacated repo to work with nixpkgs-review, it always fails on the nix-env list packages out step03:00:31
@raf:notashelf.devraf Are there any restrictions to which packages I may upgrade in cases such as https://github.com/NixOS/nixpkgs/pull/350549?

Should I stick to my own package, or is it fair to simply tackle whichever package I can? 		03:47:25
@artturin:matrix.orgArtturin
In reply to @raf:notashelf.dev
Are there any restrictions to which packages I may upgrade in cases such as https://github.com/NixOS/nixpkgs/pull/350549?

Should I stick to my own package, or is it fair to simply tackle whichever package I can?
Feel free 		03:51:42
@raf:notashelf.devraf Cheers. 03:51:54
@raf:notashelf.devraf One more thing, can I also "modernize" (format, remove `with lib;`, etc.) the derivations while I'm at it or should I open a new PR for each?

Since they are relatively minor changes, it makes more sense to get them done and dusted while I'm at it but I'd like to know what's the generally accepted way of doing those. 		03:55:21
@sigmasquadron:matrix.orgSigmaSquadron
In reply to @raf:notashelf.dev
One more thing, can I also "modernize" (format, remove `with lib;`, etc.) the derivations while I'm at it or should I open a new PR for each?

Since they are relatively minor changes, it makes more sense to get them done and dusted while I'm at it but I'd like to know what's the generally accepted way of doing those.
make a new commit for each step of those, but it's fine to leave them in the same PR. 		03:56:46
@raf:notashelf.devraf Got it, thanks. 03:57:07
@artturin:matrix.orgArtturinI'd wait for the treewide reformat to not clutter the blame or the blame file 04:00:43
@artturin:matrix.orgArtturin*if not doing significant edits04:00:54
@raf:notashelf.devraf Looks like the package I'm working on doesn't work with newer Electron versions, so scratch that. I'll keep it in mind for my future PRs. 04:07:06
@imincik:matrix.orgIvan Mincik (imincik) joined the room.07:29:25
@imincik:matrix.orgIvan Mincik (imincik) Hi what strictDeps does ? It is not properly documented . Thanks 07:30:42
@opna2608:matrix.orgPuna more or less enforces strict separation of native*Inputs and non-native *Inputs, similar to what you'd run into during cross-compilation. i.e. binaries from buildInputs are not available on PATH 07:36:00
@opna2608:matrix.orgPuna try it, see if smth breaks in a derivation. if it doesn't, great - if all its deps can be built under cross, then it's likely that at least its input separation is also correct for cross. if it does, then either some dependencies are in the wrong inputs/deps list, or they need to be put into multiple of those lists (pkg-config for build-time tools ~= pkg-config for run-time libraries), or the build manager setup is just borked with strictDeps (like Qmake TvT) 07:53:32
@emilazy:matrix.orgemily
In reply to @julius:mtx.liftm.de
Btw, what's the policy for marking a package as broken on master? guickgui has been broken for 8 weeks, there's no maintainer reaction to the related issue for 4, and it keeps annoying me by popping up in my nixpkgs-reviewsā€¦ Is that enough?
feel free, though there's always the chance someone might fix it for ZHF 		08:24:47
@emilazy:matrix.orgemily
In reply to @colemickens:matrix.org
emily: do you want to block on your comment? https://github.com/NixOS/nixpkgs/pull/350153 I think I get the sentiment, but also, do we add checks for expected files for every package? I think now that it's moved to meson, its unlikely to happen again.
not intending to block. I was hoping more for a functionality test in passthru or similar, but it's ok. there is an installCheckPhase with file existence tests now which is better than nothing. I am a bit worried about the --enable-serialconfdir=${placeholder "out"}/etc/reader.conf.d" thing ā€“ are we sure that without the configuration it is going to look there rather than the global /etc? 		08:25:52
@luna-null:matrix.orgluna-null changed their display name from Autumn to luna-null.09:49:23
@emilazy:matrix.orgemily Alyssa Ross: sorry to ping but feedback on https://github.com/NixOS/nixpkgs/pull/349555#issuecomment-2426754636 would be helpful as I'm worried about Rust bindgen regressing on Darwin for the whole 24.11 release cycle 10:18:14
@emilazy:matrix.orgemily cc Tristan Ross too to make sure this is okay freeze-wise (it should be irrelevant for the standard wrapped use of Clang in Nix derivations) 10:18:42
@emilazy:matrix.orgemily do we have any supported way of running updateScripts outside of Nixpkgs? 10:49:03
@emilazy:matrix.orgemily nix-update can't do it, it looks for ./maintainers/scripts/update.nix 10:49:13

Show newer messages

Back to Room ListRoom Version: 6