fetchgit: take `postCheckout` to allow gathering revision and commit metadata without leaving `.git` by ShamrockLee · Pull Request #465497 · NixOS/nixpkgs

ShamrockLee · 2025-11-27T04:57:33Z

Description

This PR introduces the postCheckout argument for fetchgit to allow running shell commands (especially git describe) before the .git directory is removed or sanitized. It uses nix-prefetch-git's existing support to environment variable NIX_PREFETCH_GIT_CHECKOUT_HOOK, while giving it a more friendly name.

To make git describe work when specifying fetchgit { tag = "my_tag"; ... }, this PR also creates a reference of the same name (path) as the --rev <ref>'s <ref> operand.

Partially addresses PR #8567

fetchgit with leaveDotGit = true is still not completely deterministic #8567

Things done

Add a 👍 reaction to pull requests you find important.

ShamrockLee · 2025-11-27T07:31:38Z

`nixpkgs-review` result

Generated using nixpkgs-review-gha

Command: nixpkgs-review pr 465497
Commit: 143961bc02f6973460b825bc065d65e44ad304c9 (subsequent changes)
Merge: 9ed9b097d571f324f48fea1c84c80ecab055fa47

Logs: https://github.com/ShamrockLee/nixpkgs-review-gha/actions/runs/19728343354

`x86_64-linux`

✅ 30 packages built:

tests.fetchFromBitbucket.withEncodedWhitespaceGit
tests.fetchFromGitHub.fetchTags
tests.fetchFromGitHub.leave-git
tests.fetchFromGitHub.rootDir
tests.fetchFromGitHub.sparseCheckout
tests.fetchFromGitHub.sparseCheckoutNonConeMode
tests.fetchFromGitHub.submodule-deep
tests.fetchFromGitHub.submodule-leave-git
tests.fetchFromGitHub.submodule-leave-git-deep
tests.fetchFromGitHub.submodule-simple
tests.fetchgit.dumb-http-signed-tag
tests.fetchgit.fetchTags
tests.fetchgit.git-describe-without-dot-git
tests.fetchgit.leave-git
tests.fetchgit.prefetch-git-no-add-path
tests.fetchgit.rootDir
tests.fetchgit.simple
tests.fetchgit.sparseCheckout
tests.fetchgit.sparseCheckoutNonConeMode
tests.fetchgit.submodule-deep
tests.fetchgit.submodule-leave-git
tests.fetchgit.submodule-leave-git-deep
tests.fetchgit.submodule-simple
tests.fetchgit.withGitConfig
tests.fetchtorrent.http-link (tests.fetchtorrent.http-link-transmission)
tests.fetchtorrent.magnet-link (tests.fetchtorrent.magnet-link-transmission)
tests.fetchurl.header
tests.haskell.cabalSdist.assumptionLocalHasDirectReference
tests.haskell.cabalSdist.localHasNoDirectReference
tests.testers.runCommand.bork

`aarch64-linux`

✅ 28 packages built:

tests.fetchFromBitbucket.withEncodedWhitespaceGit
tests.fetchFromGitHub.fetchTags
tests.fetchFromGitHub.leave-git
tests.fetchFromGitHub.rootDir
tests.fetchFromGitHub.sparseCheckout
tests.fetchFromGitHub.sparseCheckoutNonConeMode
tests.fetchFromGitHub.submodule-deep
tests.fetchFromGitHub.submodule-leave-git
tests.fetchFromGitHub.submodule-leave-git-deep
tests.fetchFromGitHub.submodule-simple
tests.fetchgit.dumb-http-signed-tag
tests.fetchgit.fetchTags
tests.fetchgit.git-describe-without-dot-git
tests.fetchgit.leave-git
tests.fetchgit.prefetch-git-no-add-path
tests.fetchgit.rootDir
tests.fetchgit.simple
tests.fetchgit.sparseCheckout
tests.fetchgit.sparseCheckoutNonConeMode
tests.fetchgit.submodule-deep
tests.fetchgit.submodule-leave-git
tests.fetchgit.submodule-leave-git-deep
tests.fetchgit.submodule-simple
tests.fetchgit.withGitConfig
tests.fetchtorrent.http-link (tests.fetchtorrent.http-link-transmission)
tests.fetchtorrent.magnet-link (tests.fetchtorrent.magnet-link-transmission)
tests.fetchurl.header
tests.testers.runCommand.bork

`x86_64-darwin` (sandbox = relaxed)

❌ 1 package failed to build:

tests.testers.runCommand.bork

✅ 27 packages built:

tests.fetchFromBitbucket.withEncodedWhitespaceGit
tests.fetchFromGitHub.fetchTags
tests.fetchFromGitHub.leave-git
tests.fetchFromGitHub.rootDir
tests.fetchFromGitHub.sparseCheckout
tests.fetchFromGitHub.sparseCheckoutNonConeMode
tests.fetchFromGitHub.submodule-deep
tests.fetchFromGitHub.submodule-leave-git
tests.fetchFromGitHub.submodule-leave-git-deep
tests.fetchFromGitHub.submodule-simple
tests.fetchgit.dumb-http-signed-tag
tests.fetchgit.fetchTags
tests.fetchgit.git-describe-without-dot-git
tests.fetchgit.leave-git
tests.fetchgit.prefetch-git-no-add-path
tests.fetchgit.rootDir
tests.fetchgit.simple
tests.fetchgit.sparseCheckout
tests.fetchgit.sparseCheckoutNonConeMode
tests.fetchgit.submodule-deep
tests.fetchgit.submodule-leave-git
tests.fetchgit.submodule-leave-git-deep
tests.fetchgit.submodule-simple
tests.fetchgit.withGitConfig
tests.fetchtorrent.http-link (tests.fetchtorrent.http-link-transmission)
tests.fetchtorrent.magnet-link (tests.fetchtorrent.magnet-link-transmission)
tests.fetchurl.header

Error logs: `x86_64-darwin`

tests.testers.runCommand.bork

                    exc.stdout, exc.stderr = process.communicate()
                else:
                    # POSIX _communicate already populated the output so
                    # far into the TimeoutExpired exception.
                    process.wait()
                raise
            except:  # Including KeyboardInterrupt, communicate handled that.
                process.kill()
                # We don't call process.wait() as .__exit__ does that for us.
                raise
            retcode = process.poll()
            if check and retcode:
>               raise CalledProcessError(retcode, process.args,
                                         output=stdout, stderr=stderr)
E               subprocess.CalledProcessError: Command '['/nix/store/84s01mzl74vh69aphpl3rf07pp49pkz0-python3-3.13.9/bin/python3.13', '-m', 'bork', 'download', 'gh:duckinator/emanate', 'v7.0.0']' returned non-zero exit status 1.
/nix/store/84s01mzl74vh69aphpl3rf07pp49pkz0-python3-3.13.9/lib/python3.13/subprocess.py:577: CalledProcessError

=========================== short test summary info ============================

FAILED bork/tests/test_cmd_download.py::test_download - subprocess.CalledProcessError: Command '['/nix/store/84s01mzl74vh69aphpl3rf...

============= 1 failed, 3 passed, 5 deselected in 81.45s (0:01:21) =============

`aarch64-darwin` (sandbox = relaxed)

✅ 28 packages built:

tests.fetchFromBitbucket.withEncodedWhitespaceGit
tests.fetchFromGitHub.fetchTags
tests.fetchFromGitHub.leave-git
tests.fetchFromGitHub.rootDir
tests.fetchFromGitHub.sparseCheckout
tests.fetchFromGitHub.sparseCheckoutNonConeMode
tests.fetchFromGitHub.submodule-deep
tests.fetchFromGitHub.submodule-leave-git
tests.fetchFromGitHub.submodule-leave-git-deep
tests.fetchFromGitHub.submodule-simple
tests.fetchgit.dumb-http-signed-tag
tests.fetchgit.fetchTags
tests.fetchgit.git-describe-without-dot-git
tests.fetchgit.leave-git
tests.fetchgit.prefetch-git-no-add-path
tests.fetchgit.rootDir
tests.fetchgit.simple
tests.fetchgit.sparseCheckout
tests.fetchgit.sparseCheckoutNonConeMode
tests.fetchgit.submodule-deep
tests.fetchgit.submodule-leave-git
tests.fetchgit.submodule-leave-git-deep
tests.fetchgit.submodule-simple
tests.fetchgit.withGitConfig
tests.fetchtorrent.http-link (tests.fetchtorrent.http-link-transmission)
tests.fetchtorrent.magnet-link (tests.fetchtorrent.magnet-link-transmission)
tests.fetchurl.header
tests.testers.runCommand.bork

ShamrockLee · 2025-11-27T08:16:29Z

tests.testers.runCommand.bork is known to be flaky, and is unrelated to fetchgit.

ShamrockLee · 2025-11-27T10:20:35Z

Test cases improved.

bjornfor · 2025-11-27T15:06:24Z

Be aware that git describe is not a reproducible command. Cloning a repo at two different times can produce different git describe results even if the working tree is checked out at the same commit. So there is a risk associated with caching the command output.

Atemu

This sounds and looks sensible to me

ShamrockLee · 2025-11-27T23:52:15Z

Be aware that git describe is not a reproducible command. Cloning a repo at two different times can produce different git describe results even if the working tree is checked out at the same commit.

Is there a reproducer to demonstrate this non-reproducibility?

Atemu · 2025-11-28T02:31:50Z

Also, wouldn't this only occur when upstream changes tags retroactively? That's not a case our regular fetchgit can handle gracefully either.

bjornfor · 2025-11-28T06:28:58Z

Be aware that git describe is not a reproducible command. Cloning a repo at two different times can produce different git describe results even if the working tree is checked out at the same commit.

Is there a reproducer to demonstrate this non-reproducibility?

Create repo with commit abcd123 and tag 1.0-rc1 at that commit.
Fetch repo at commit abc123. Capture the hash of it (and the git describe output) in Nix.
The next day, someone pushes a new tag on the same commit: 1.0
Fetches since then will be rejected by Nix, since git describe returns a different output and the hash does not match anymore.

bjornfor · 2025-11-28T06:34:08Z

Also, wouldn't this only occur when upstream changes tags retroactively?

Not only change. It can happen if they add a tag anywhere that git describe finds quicker than the previous tag (EDIT: s/commit/tag/).

That's not a case our regular fetchgit can handle gracefully either.

Right, if you specify fetchgit { tag = ...; } instead of {fetchgit { rev = ...; }. (Hm, I haven't though about that before.)

me-and · 2025-11-28T14:35:36Z

I'm not worried about the non-determinism of git describe here: fundamentally we're always accessing something from a third party, so the cast-iron guarantee that things haven't changed unexpectedly is the FOD hash not changing.

It's on the user to make sure they use the tools in a way that isn't inherently fragile. Someone relying on a source package marked "development latest" would be a poor idea, as would relying on a Git repository where tags aren't fixed (as GitHub recommends with GitHub Actions, for example).

I'd expect using git describe to be a bad idea in some circumstances, and a good idea in others. As a key example, I use it in https://github.com/me-and/nix-git-ci, which is testing the Git source code itself, where (a) unexpected changes to the describe output wouldn't be a problem and (b) the Git source repository has clear policies that make that unlikely anyway.

pkgs/build-support/fetchgit/tests.nix

ShamrockLee · 2025-11-30T07:47:42Z

Be aware that git describe is not a reproducible command. Cloning a repo at two different times can produce different git describe results even if the working tree is checked out at the same commit.

Is there a reproducer to demonstrate this non-reproducibility?
1. Create repo with commit abcd123 and tag 1.0-rc1 at that commit.

2. Fetch repo at commit abc123. Capture the hash of it (and the git describe output) in Nix.

3. The next day, someone pushes a new tag on the same commit: 1.0

4. Fetches since then will be rejected by Nix, since `git describe` returns a different output and the hash does not match anymore.

So the non-reproducibility is brought by fetchTags ~~or deepClone~~, and such non-reproducibility might affect hash stability by the use of postFetch or postCheckout.

Considering this, let's avoid the use of fetchTags, but name the tag reference as its original name (refs/tags/*) instead of just FETCH_HEAD. If applied, git describe will work deterministically inside postCheckout when specifying fetchgit { tags = ...; }.

As for unstable versions that specifies the Git revision directly but still want git describe to describe the last tag, we could introduce a new attribute extraTags or make fetchTags to optionally take a list. (The latter requires __structuredAttrs = true that PR #464475 introduces.)

Cc: @bjornfor @me-and

ShamrockLee · 2025-12-10T13:34:12Z

All the tests.fetchgit tests now build on my x86_64 machine.

philiptaron · 2025-12-10T16:02:53Z

I'd like to merge in a day or two. Hopefully @me-and has a chance to review.

me-and · 2025-12-14T20:36:26Z

I'd like to merge in a day or two. Hopefully @me-and has a chance to review.

I should be able to take a look (and, more to the point, a think) in the next couple of days. If you want to merge before then, that's absolutely fine; if I have any feedback that really needs addressing we can iterate from that point.

me-and

This looks fantastically useful, thank you! I think it wants a few fixes before it's quite ready to be merged, but it's all just tidying to improve long-term maintainability; the fundamental changes are spot on.

`nixpkgs-review` result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 465497
Commit: 29458868f15aa1d1227e931f2c9f07690ba2fc5e

`x86_64-linux`

✅ 92 packages built:

bundix
cabal2nix
chirpstack-concentratord
corto
crate2nix
crush
crystal2nix
dub-to-nix
gcalcli
gcalcli.dist
git-unroll
haskellPackages.cli-nix
haskellPackages.cli-nix.doc
haskellPackages.nix-thunk
haskellPackages.nix-thunk.doc
haskellPackages.nvfetcher
haskellPackages.nvfetcher.doc
haskellPackages.update-nix-fetchgit
haskellPackages.update-nix-fetchgit.doc
kcl
lixPackageSets.git.nix-update
lixPackageSets.git.nix-update.dist
lixPackageSets.latest.nix-update (lixPackageSets.lix_2_94.nix-update)
lixPackageSets.latest.nix-update.dist (lixPackageSets.lix_2_94.nix-update.dist)
lixPackageSets.stable.nix-update (lixPackageSets.lix_2_93.nix-update)
lixPackageSets.stable.nix-update.dist (lixPackageSets.lix_2_93.nix-update.dist)
localtunnel
lon
luarocks-packages-updater
luarocks-packages-updater.dist
mdwatch
meshlab
mlv-app
nim_lk
nix-prefetch-git
nix-prefetch-scripts
nix-update
nix-update-source
nix-update-source.dist
nix-update.dist
npins
nvfetcher
outline
prefetch-yarn-deps
prowlarr
python312Packages.pymeshlab
python313Packages.nixpkgs-updaters-library
python313Packages.nixpkgs-updaters-library.dist
python313Packages.pymeshlab
radarr
sonarr
sus-compiler
tests.fetchFromBitbucket.withEncodedWhitespaceGit
tests.fetchFromGitHub.fetchTags
tests.fetchFromGitHub.leave-git
tests.fetchFromGitHub.rootDir
tests.fetchFromGitHub.sparseCheckout
tests.fetchFromGitHub.sparseCheckoutNonConeMode
tests.fetchFromGitHub.submodule-deep
tests.fetchFromGitHub.submodule-leave-git
tests.fetchFromGitHub.submodule-leave-git-deep
tests.fetchFromGitHub.submodule-simple
tests.fetchgit.cached-prefetch-avoids-fetch
tests.fetchgit.collect-rev
tests.fetchgit.describe-tag
tests.fetchgit.dumb-http-signed-tag
tests.fetchgit.fetchTags
tests.fetchgit.leave-git
tests.fetchgit.prefetch-git-no-add-path
tests.fetchgit.rootDir
tests.fetchgit.simple
tests.fetchgit.simple-tag
tests.fetchgit.sparseCheckout
tests.fetchgit.sparseCheckoutNonConeMode
tests.fetchgit.submodule-deep
tests.fetchgit.submodule-leave-git
tests.fetchgit.submodule-leave-git-deep
tests.fetchgit.submodule-revision-count
tests.fetchgit.submodule-simple
tests.fetchgit.withGitConfig
tests.fetchurl.header
tests.haskell.cabalSdist.assumptionLocalHasDirectReference
tests.haskell.cabalSdist.localHasNoDirectReference
tests.testers.runCommand.bork
typescript-language-server
update-nix-fetchgit
update-python-libraries
vclab-nexus
vclab-nexus.bin
vimPluginsUpdater
xosd-xft
yarn2nix

pkgs/build-support/fetchgit/tests.nix

pkgs/build-support/fetchgit/nix-prefetch-git

…T_HOOK Co-authored-by: Adam Dinwoodie <adam@dinwoodie.org>

Co-authored-by: Adam Dinwoodie <adam@dinwoodie.org>

me-and

Fantastic. Thank you!

RaitoBezarius · 2025-12-23T17:03:50Z

Commit 7e08567 broke nix build git+https://git.afnix.fr/afnix/nix-gerrit#gerrit.src --rebuild --override-input nixpkgs "github:nixos/nixpkgs?rev=7e085677f9961b1bf06f292198614e2bdeede9d4, was it intended?

It seems like this commit is not achieving its intended effect as tags are indeed not fetched where it was expected before.

This should probably be reverted as it's breaking backward compatibility?

ShamrockLee · 2025-12-24T00:38:19Z

Commit 7e08567 broke nix build git+https://git.afnix.fr/afnix/nix-gerrit#gerrit.src --rebuild --override-input nixpkgs "github:nixos/nixpkgs?rev=7e085677f9961b1bf06f292198614e2bdeede9d4, was it intended

This is not intended.

@RaitoBezarius Could you help test if PR #473411 now restores the hash compatibility?

Following NixOS#465497 it's possible to specify `postCheckout` to perform actions after repo is fetched but before `.git` is deleted. Signed-off-by: Marcin Serwin <marcin@serwin.dev>

nixpkgs-ci bot added 10.rebuild-linux: 11-100 This PR causes between 11 and 100 packages to rebuild on Linux. 10.rebuild-darwin: 11-100 This PR causes between 11 and 100 packages to rebuild on Darwin. 6.topic: fetch Fetchers (e.g. fetchgit, fetchsvn, ...) labels Nov 27, 2025

ShamrockLee mentioned this pull request Nov 27, 2025

fetchgit with leaveDotGit = true is still not completely deterministic #8567

Open

ShamrockLee marked this pull request as ready for review November 27, 2025 08:14

nixpkgs-ci bot requested a review from philiptaron November 27, 2025 08:14

ShamrockLee force-pushed the fetchgit-postCheckout branch from 143961b to 08c189f Compare November 27, 2025 10:18

ShamrockLee requested review from Atemu, eclairevoyant, me-and and tobiasBora November 27, 2025 10:27

ShamrockLee force-pushed the fetchgit-postCheckout branch from 08c189f to 63c5be3 Compare November 27, 2025 10:36

ShamrockLee changed the title ~~fetchgit: take postCheckout to allow git-describe without leaving .git~~ fetchgit: take postCheckout to allow gathering commit messages without leaving .git Nov 27, 2025

ShamrockLee changed the title ~~fetchgit: take postCheckout to allow gathering commit messages without leaving .git~~ fetchgit: take postCheckout to allow gathering revision and commit metadata without leaving .git Nov 27, 2025

Atemu reviewed Nov 27, 2025

View reviewed changes

me-and suggested changes Nov 28, 2025

View reviewed changes

pkgs/build-support/fetchgit/tests.nix Outdated Show resolved Hide resolved

pkgs/build-support/fetchgit/tests.nix Outdated Show resolved Hide resolved

ShamrockLee force-pushed the fetchgit-postCheckout branch 3 times, most recently from 8d97c10 to 65f011e Compare November 30, 2025 07:38

ShamrockLee requested a review from me-and November 30, 2025 07:47

ShamrockLee force-pushed the fetchgit-postCheckout branch from eb3c6ad to 2945886 Compare December 10, 2025 14:24

philiptaron approved these changes Dec 10, 2025

View reviewed changes

nixpkgs-ci bot added the 12.approvals: 1 This PR was reviewed and approved by one person. label Dec 10, 2025

me-and suggested changes Dec 16, 2025

View reviewed changes

nix-prefetch-git: fetch the fetching tag for NIX_PREFETCH_GIT_CHECKOU…

d4dd261

…T_HOOK Co-authored-by: Adam Dinwoodie <adam@dinwoodie.org>

ShamrockLee force-pushed the fetchgit-postCheckout branch from 2945886 to e348c7b Compare December 16, 2025 17:16

ShamrockLee and others added 2 commits December 17, 2025 01:19

nix-prefetch-git: dont't fetch tags when deep clone unless leaving .git

7e08567

Co-authored-by: Adam Dinwoodie <adam@dinwoodie.org>

tests.fetchgit: add submodule-revision-count

8974669

ShamrockLee force-pushed the fetchgit-postCheckout branch from e348c7b to 8974669 Compare December 16, 2025 17:19

ShamrockLee requested a review from me-and December 16, 2025 17:20

me-and approved these changes Dec 17, 2025

View reviewed changes

nixpkgs-ci bot added 12.approvals: 2 This PR was reviewed and approved by two persons. and removed 12.approvals: 1 This PR was reviewed and approved by one person. labels Dec 17, 2025

philiptaron approved these changes Dec 17, 2025

View reviewed changes

philiptaron added this pull request to the merge queue Dec 17, 2025

Merged via the queue into NixOS:master with commit cca4ec8 Dec 17, 2025
31 of 34 checks passed

Aleksanaa mentioned this pull request Dec 21, 2025

shipwright: enable fetchTags #473003

Merged

13 tasks

me-and mentioned this pull request Dec 22, 2025

nix-prefetch-git: cannot fetch branch when using --deepClone #473332

Closed

3 tasks

RaitoBezarius mentioned this pull request Dec 23, 2025

Revert "fetchgit: take postCheckout to allow gathering revision and commit metadata without leaving .git" #473619

Closed

ShamrockLee mentioned this pull request Dec 24, 2025

nix-prefetch-git: fix tag fetching and restore .git hash compatibility for postCheckout PR #473411

Merged

13 tasks

ShamrockLee mentioned this pull request Dec 29, 2025

nix-prefetch-git: fix checkout_ref() for "dumb http" repositories #116289

Closed

10 tasks

marcin-serwin mentioned this pull request Jan 5, 2026

fetchFromGithub: use fetchgit when postCheckout is not an empty string #477213

Closed

13 tasks

Uh oh!

Conversation

ShamrockLee commented Nov 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Things done

Uh oh!

ShamrockLee commented Nov 27, 2025

nixpkgs-review result

x86_64-linux

aarch64-linux

x86_64-darwin (sandbox = relaxed)

aarch64-darwin (sandbox = relaxed)

Uh oh!

ShamrockLee commented Nov 27, 2025

Uh oh!

ShamrockLee commented Nov 27, 2025

Uh oh!

bjornfor commented Nov 27, 2025

Uh oh!

Atemu left a comment

Choose a reason for hiding this comment

Uh oh!

ShamrockLee commented Nov 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Atemu commented Nov 28, 2025

Uh oh!

bjornfor commented Nov 28, 2025

Uh oh!

bjornfor commented Nov 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

me-and commented Nov 28, 2025

Uh oh!

Uh oh!

Uh oh!

ShamrockLee commented Nov 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ShamrockLee commented Dec 10, 2025

Uh oh!

philiptaron commented Dec 10, 2025

Uh oh!

me-and commented Dec 14, 2025

Uh oh!

me-and left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

nixpkgs-review result

x86_64-linux

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

me-and left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

RaitoBezarius commented Dec 23, 2025

Uh oh!

ShamrockLee commented Dec 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

ShamrockLee commented Nov 27, 2025 •

edited

Loading

`nixpkgs-review` result

`x86_64-linux`

`aarch64-linux`

`x86_64-darwin` (sandbox = relaxed)

`aarch64-darwin` (sandbox = relaxed)

ShamrockLee commented Nov 27, 2025 •

edited

Loading

bjornfor commented Nov 28, 2025 •

edited

Loading

ShamrockLee commented Nov 30, 2025 •

edited

Loading

me-and left a comment •

edited

Loading

`nixpkgs-review` result

`x86_64-linux`

ShamrockLee commented Dec 24, 2025 •

edited

Loading