Skip to content

Bump djangorestframework-stubs from 3.17.1 to 3.18.0 - #2260

Open
dependabot[bot] wants to merge 1 commit into
mainfrom
dependabot/pip/djangorestframework-stubs-3.18.0
Open

Bump djangorestframework-stubs from 3.17.1 to 3.18.0#2260
dependabot[bot] wants to merge 1 commit into
mainfrom
dependabot/pip/djangorestframework-stubs-3.18.0

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Aug 17, 2026

Copy link
Copy Markdown
Contributor

Bumps djangorestframework-stubs from 3.17.1 to 3.18.0.

Release notes

Sourced from djangorestframework-stubs's releases.

3.18.0

  • This is the first version to be based on DRF 3.18
  • This is the last version to support django-stubs 6.0.x - next release will require 6.1.x
  • This is the last version to support Python 3.10 - following django-stubs upstream

What's Changed

Housekeeping

Full Changelog: typeddjango/djangorestframework-stubs@3.17.1...3.18.0

Commits

Dependabot compatibility score

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.


Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
  • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

Bumps [djangorestframework-stubs](https://github.com/typeddjango/djangorestframework-stubs) from 3.17.1 to 3.18.0.
- [Release notes](https://github.com/typeddjango/djangorestframework-stubs/releases)
- [Commits](typeddjango/djangorestframework-stubs@3.17.1...3.18.0)

---
updated-dependencies:
- dependency-name: djangorestframework-stubs
  dependency-version: 3.18.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file python Pull requests that update Python code labels Aug 17, 2026
@claude

claude Bot commented Aug 17, 2026

Copy link
Copy Markdown

Review

This is a routine Dependabot bump of djangorestframework-stubs (dev-only mypy type stubs) from 3.17.1 → 3.18.0 in requirements/local.txt, tracking the corresponding DRF 3.18.0 stub release. Single line diff, no production code touched.

Compatibility check: 3.18.0 is noted upstream as "the last version to support django-stubs 6.0.x" — this repo pins django-stubs==6.0.7, so we're still inside the supported range. No action needed there, but worth remembering that the next stubs bump will likely require moving django-stubs to 6.1.x in lockstep.

Risk assessment:

  • No runtime/production impact — this package only affects mypy type checking in local/dev.
  • No test coverage concerns — nothing here is exercised by the test suite; correctness is verified by mypy passing in CI/pre-commit.
  • No security concerns.
  • Per CLAUDE.md's "no dead code" / cleanliness guidance, worth a quick sanity check that mypy still passes cleanly with the updated stubs (the upstream changelog mentions dropped coreapi/coreschema stubs, which were already removed from DRF itself in 3.17+, so this should be a no-op for this codebase unless those symbols are still referenced somewhere).

Looks safe to merge once CI (mypy/lint) is green.

JSv4 added a commit that referenced this pull request Aug 20, 2026
…ired

`main`'s branch protection has no `required_status_checks` object at all, so
nothing gates a merge on CI having run, let alone passed. PR #2262 merged with
Backend CI never having run on its head commit at all -- and because nothing
was required, "no check reported" was not a blocker. The push that merged it
then failed at the linter, which skipped `pytest` (0s); `main` sat that way for
~30 hours, repaired only by accident when an unrelated PR's
`pre-commit run --all-files` happened to reformat the same file.

Requiring *something* is therefore the fix, but requiring the `pytest` job is
not, because it leaves a second hole open and opens a third:

  * GitHub reports a job skipped by its own `if:` as SUCCESS to branch
    protection. `pytest` is gated on `needs.linter.result == 'success'`, so a
    red linter skips it and a required `pytest` still reads green. This is not
    hypothetical: PRs #2260, #2264 and #2265 are all sitting at
    `linter=failure / pytest=skipped` right now, and would be mergeable under
    that policy with a red linter.
  * A workflow skipped by path filtering never reports its checks at all, so
    the required check hangs Pending forever. With `paths-ignore: docs/**` on
    the `pull_request` trigger, requiring any job here would make docs-only
    PRs permanently unmergeable.

So the requirable check has to always run and inspect the other jobs itself,
telling "skipped because this PR touches no backend code" apart from "skipped
because something upstream broke". That is the new `gate` job; its decision
table is `.github/scripts/backend_ci_gate.sh`, which carries a `--self-test`
that the job runs on every invocation -- a gate whose own logic has silently
inverted is worse than no gate.

`paths-ignore` is dropped from the `pull_request` trigger for the reason
above; the `changes` path filter still keeps the expensive jobs from running,
so a docs-only PR now costs two ubuntu-latest jobs of a few seconds.

`require_backend_ci_gate.sh` applies the protection change itself, because the
obvious `gh api` call is a footgun: `PUT .../branches/main/protection` replaces
the ENTIRE object (dropping review rules and the force-push/deletion bans
unless they are re-sent), and the narrower
`PATCH .../protection/required_status_checks` sub-resource 404s when no such
object exists yet. It refuses to require a context name that has never been
reported on the branch, since that would block every PR with no error anywhere.

Verified by replaying the gate over the last 60 Backend CI runs: it blocks all
8 PR runs with a red linter and both of the merge-commit runs from #2262's
window, and allows all 25 genuinely green runs and the 3 with no backend
changes.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
JSv4 added a commit that referenced this pull request Aug 20, 2026
…ired

`main`'s branch protection has no `required_status_checks` object at all, so
nothing gates a merge on CI having run, let alone passed. PR #2262 merged with
Backend CI never having run on its head commit at all -- and because nothing
was required, "no check reported" was not a blocker. The push that merged it
then failed at the linter, which skipped `pytest` (0s); `main` sat that way for
~30 hours, repaired only by accident when an unrelated PR's
`pre-commit run --all-files` happened to reformat the same file.

Requiring *something* is therefore the fix, but requiring the `pytest` job is
not, because it leaves a second hole open and opens a third:

  * GitHub reports a job skipped by its own `if:` as SUCCESS to branch
    protection. `pytest` is gated on `needs.linter.result == 'success'`, so a
    red linter skips it and a required `pytest` still reads green. This is not
    hypothetical: PRs #2260, #2264 and #2265 are all sitting at
    `linter=failure / pytest=skipped` right now, and would be mergeable under
    that policy with a red linter.
  * A workflow skipped by path filtering never reports its checks at all, so
    the required check hangs Pending forever. With `paths-ignore: docs/**` on
    the `pull_request` trigger, requiring any job here would make docs-only
    PRs permanently unmergeable.

So the requirable check has to always run and inspect the other jobs itself,
telling "skipped because this PR touches no backend code" apart from "skipped
because something upstream broke". That is the new `gate` job; its decision
table is `.github/scripts/backend_ci_gate.sh`, which carries a `--self-test`
that the job runs on every invocation -- a gate whose own logic has silently
inverted is worse than no gate.

`paths-ignore` is dropped from the `pull_request` trigger for the reason
above; the `changes` path filter still keeps the expensive jobs from running,
so a docs-only PR now costs two ubuntu-latest jobs of a few seconds.

`require_backend_ci_gate.sh` applies the protection change itself, because the
obvious `gh api` call is a footgun: `PUT .../branches/main/protection` replaces
the ENTIRE object (dropping review rules and the force-push/deletion bans
unless they are re-sent), and the narrower
`PATCH .../protection/required_status_checks` sub-resource 404s when no such
object exists yet. It refuses to require a context name that has never been
reported on the branch, since that would block every PR with no error anywhere.

Verified by replaying the gate over the last 60 Backend CI runs: it blocks all
8 PR runs with a red linter and both of the merge-commit runs from #2262's
window, and allows all 25 genuinely green runs and the 3 with no backend
changes.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file python Pull requests that update Python code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants