7 Commits
Author SHA1 Message Date
William Woodruff b0fdc328c6 Update CI Node.js to 24.22.1 for the eMudhra fix (#1146)
<details>
<summary>Automated notes</summary>

Use Node.js 24.22.1 for the `setup-node` jobs affected by the eMudhra
intermediate CA compromise. [Node.js security
release](https://nodejs.org/en/blog/vulnerability/october-2026-security-releases)

Both workflow jobs read their Node.js version from `.nvmrc`.

</details>
2026-11-17 15:44:55 -05:00
William Woodruff de938e707d Add the require-checksums input (#1139)
Adds `require-checksums`, which enforces the "intrinsic or explicitly
provided checksum" idea. See #1032.

<details>
<summary>Automated notes</summary>

Add the opt-in input `require-checksums`. When enabled, setup-uv
verifies the archive against the workflow's `checksum` input or the
digest bundled for the resolved uv version, architecture, and platform.
If neither is available, setup fails before downloading the archive. By
default, setup-uv continues to accept manifest checksums as it does
today.

The setting also applies to custom manifests, which can still supply
download URLs. On persistent runners, strict mode keys tool-cache
entries by the expected archive digest so a previous manifest-verified
installation cannot bypass the check.

Closes #1032. Builds on #1033.

## Validation

- `npm ci --ignore-scripts` and `npm run all` passed under Node 24 (182
tests).
- The packaged-action regression test installs a custom build into a
temporary runner tool cache. It checks that strict mode rejects a
manifest-only version even with a warm cache, reuses a verified
installation for the same digest, and rejects a changed digest or a
custom build that conflicts with the bundled digest.

</details>

---------

Signed-off-by: William Woodruff <william@yossarian.net>
2026-11-16 22:24:53 +01:00
William Woodruff fc0acc2d6f Pin astral-sh/github-actions to 2026.11.16.0 (#1145)
Pin the shared actions to commit
`67dc9ee1ddc113eccd8116e7b949fb6cdcd89889` from [release
2026.11.16.0](https://github.com/astral-sh/github-actions/releases/tag/2026.11.16.0).
Add the release version beside each YAML pin.

Parsed the changed YAML and verified that each referenced action or
reusable workflow exists at the pinned commit. The workflow also passed
actionlint and offline zizmor checks.
2026-11-16 12:10:39 -05:00
William Woodruff 68ef1d76d3 Run demudhra first on macOS (#1125) 2026-11-03 13:23:25 -05:00
William Woodruff f634bf473a Expose a Python "identity" output (#1036)
https://github.com/pyca/cryptography/pull/15572#discussion_r3913508686
has the context for this: TL;DR our current `python-version` output
mirrors the "request" version exactly, which means that it's
insufficient for any downstream that needs to manage its own cache keys
(since caches shouldn't be shared across release candidates, but the `uv
python` request version doesn't include RC numbers).

The first commit here was my attempt to fix this by exposing the runtime
Python version, but this too is imprecise: the runtime version doesn't
indicate the interpreter variant (e.g. freethreading), which is also
important to capture in the cache identity.

My solution here is to expose `python-runtime-id`, which is just the
`key` of the active Python version from `uv python list
--output-format=json`.

---------

Signed-off-by: William Woodruff <william@yossarian.net>
2026-09-09 18:45:52 +02:00
William Woodruff 7211c71869 Use JSON + a typed wrapper instead of TS codegen (#1025)
Signed-off-by: William Woodruff <william@yossarian.net>
2026-08-19 17:54:07 -04:00
William Woodruff b1836110f7 chore(ci): address CI lint findings (#545)
This addresses all of zizmor's non-pedantic findings, and adds a
workflow to proactively flag any more that come in.

Key changes:

* I've hash-pinned all actions references. Dependabot will continue to
keep these updated and will update the hash comments as well.
* I've marked every `actions/checkout` with `persist-credentials: false`
except for one that actually needs persisted credentials (which I've
explicitly enabled with an explanatory comment)
* I've dropped some workflow-level permissions in favor of job-level
permissions that were already provisioned.
* I fixed two small template injections caused by expanding output
contexts. I think these were not exploitable in practice, but fixing
them is good for defense in depth (and makes spellcheck work nicely on
these steps).

---------

Signed-off-by: William Woodruff <william@astral.sh>
2025-09-02 13:29:06 +00:00