Update bandit requirement from >=1.7 to >=1.9.4 in /services/api - #284
Update bandit requirement from >=1.7 to >=1.9.4 in /services/api#284dependabot[bot] wants to merge 1 commit into
Conversation
Updates the requirements on [bandit](https://github.com/PyCQA/bandit) to permit the latest version. - [Release notes](https://github.com/PyCQA/bandit/releases) - [Commits](PyCQA/bandit@1.7.0...1.9.4) --- updated-dependencies: - dependency-name: bandit dependency-version: 1.9.4 dependency-type: direct:development ... Signed-off-by: dependabot[bot] <support@github.com>
|
Strix is installed on this repository, but we couldn't run this PR security review because this workspace's trial has ended. Add a card to resume code reviews here. So far, Strix has reviewed 28 pull requests, surfaced 3 security issues (1 critical/high) and blocked 2 risky merges across this workspace. |
… CI installs Five npm dev-dependency upgrades, applied together and verified as ONE state: eslint 10.8.0 -> 10.8.1, globals 17.7 -> 17.11, happy-dom 20.8.9 -> 20.11.2, typescript-eslint 8.65 -> 8.67, vite 8.1.5 -> 8.2.1 (dependabot #278 #280 #283 #285 #287). Applied in one commit rather than merged one by one because all five rewrite package-lock.json and would have conflicted pairwise. DEPENDABOT COULD NOT SEE THE WHOLE PICTURE. The ROOT workspace declares vite, eslint and happy-dom too; dependabot only opened PRs against apps/web. So merging them as-is would have changed less than they claimed: * eslint would have stayed on the root's 10.8.0 - hoisted, so the apps/web bump was a NO-OP for linting, the very tool the bump exists to update; * vite would have been installed TWICE - root's exact 8.1.5 against web's exact 8.2.1, forcing a nested second copy. Both roots are aligned; all five now resolve to single hoisted copies at the intended versions, confirmed by reading node_modules rather than the manifest. There is also an `overrides` block pinning eslint repo-wide that dependabot never touches. The first edit landed on IT instead of the devDependency - the same key lives in both sections and a count-limited replace takes whichever comes first. npm rejected it loudly (EOVERRIDE), which is the good case. Both now read 10.8.1 and are asserted equal. toolchainDocs.test.ts then caught the third thing: docs/engineering/web-standards.md still said "eslint 10.8.0". A governed doc whose versions are asserted against the manifest - and whose own comment records that both versions there were wrong for weeks once before. THE SEVEN PYTHON PRs ARE NOT MERGED HERE, deliberately. CI installs from services/api/requirements.lock with --require-hashes, NOT requirements.txt, and those PRs only raise >= floors: * #289 manifold3d, #286 fastapi, #282 shapely, #284 bandit - the lock already pins versions satisfying the new floors; safe, purely making the declaration honest. * #277 numpy, #281 trimesh, #288 boto3 - raise the floor ABOVE what the lock pins (2.5.1, 4.12.2, 1.43.46). test_lock_satisfies_requirements.py guards exactly this and was mutation-tested by simulating #281. Merging without a lock regeneration reds main - and CI would still install the old versions, so a green run would prove nothing about trimesh 5. Web typecheck + lint clean, 1651/1651, build OK. Backend 608/608 (the one failure in the first run was test_changelog_current doing its job: version bumped before this entry existed). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Four Python floor bumps applied (dependabot #289 manifold3d, #286 fastapi, #282 shapely, #284 bandit). Each raises a >= floor to a version the lock ALREADY pins, so nothing about what installs changes - the declaration simply stops understating what the project requires. test_lock_satisfies_requirements: 50 floors, 110 pins. The other three are refused. #277 numpy (lock 2.5.1), #281 trimesh (4.12.2, a major) and #288 boto3 (1.43.46) raise the floor ABOVE the lock. Two independent reasons: the lock gate would go red (mutation-tested by simulating #281, which fails naming the package, both versions and the line), and CI installs from requirements.lock with --require-hashes, so merging them would produce a green run that tested the OLD versions. A floor bump is a claim about what is required; only a lock regeneration changes what is USED. Two things the mapping turned up, both from checking rather than assuming: * services/api/requirements.txt DOES NOT EXIST. The source is requirements.in, compiled to requirements.lock. A first pass looking for the .txt found nothing and would have read as "these packages are undeclared". * shapely is declared in TWO files - requirements.in at >=2.0 and services/data/requirements.txt at >=2.1.2. Reading the wrong one makes #282 look already applied. Edits are line-anchored and assert exactly one match, so a floor like >=2.0 cannot silently match inside >=2.0.1. Backend 608/608; all five dependency gates pass (lock-satisfies, licence-lock, lock-advisories, licence-allowlist, vendor-drift). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
|
Applied in d28f8d2 (v0.3.985), together with the other three safe floor bumps in one verified commit. This floor was raised to a version Applied rather than merged so all four could be verified as one state, and because the mapping needed checking first: |
|
OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting If you change your mind, just re-open this PR and I'll resolve any conflicts on it. |
Updates the requirements on bandit to permit the latest version.
Release notes
Sourced from bandit's releases.
Commits
92ae8b8Fix B106 reporting wrong line number on multiline function calls (#1360)c8c8a55Lower version guard in check_ast_node to Python 3.12 (#1355)8f2f928Fix B615 false positive when revision is set via variable (#1358)e27493fInclude filename in nosec 'no failed test' warning (#1363)b69b336Fix B613 crash when reading from stdin (#1361)e418b79Bump docker/build-push-action from 6.18.0 to 6.19.2 (#1357)ff646fdBump docker/login-action from 3.6.0 to 3.7.0 (#1353)c0def6cchore: fixed some typos in comments (#1351)765f00dLimit B614 to torch.load deserializers (#1348)06fbbabBump docker/setup-buildx-action from 3.11.1 to 3.12.0 (#1347)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 rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill 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 versionwill 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 dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)