ci(gitlab): land .gitlab-ci.yml alongside gitea workflow (ADR-0028 phase 2) (#241)
CI / check (push) Successful in 2m38s
CI / commits (push) Has been skipped
CI / scan (push) Successful in 2m47s
CI / a11y (push) Successful in 1m29s
CI / perf (push) Successful in 4m41s

## Summary

Bump the existing `tmp` security override from `>=0.2.4` to `>=0.2.6` to cover [GHSA-ph9p-34f9-6g65](https://github.com/advisories/GHSA-ph9p-34f9-6g65) — a path-traversal vulnerability via unsanitized prefix/postfix in `tmp@<0.2.6`. Surfaced by `pnpm ci:audit` on the next CI run; both Gitea and the in-flight GitLab Phase 2 pipeline fail without this.

## Why this isn't an upstream upgrade

`tmp` is a transitive dependency. The two consuming paths today:

| Path | Latest available |
| --- | --- |
| `.>nx>tmp` | `nx@22.7.4` still declares `tmp@^0.2.4` |
| `.>@lhci/cli>tmp` | `@lhci/cli@0.15.1` (latest) still declares `tmp@^0.1.0` |

Upgrading `nx` or `@lhci/cli` does not move `tmp` to 0.2.6. The pre-existing `pnpm.overrides` entry was already pinning `tmp@<0.2.4` → `>=0.2.4` against the prior advisory; this PR just widens the lower bound to match the new one. Same pattern, one line.

## What lands

| File | Change |
| --- | --- |
| `package.json` | `"tmp@<0.2.4": ">=0.2.4"` → `"tmp@<0.2.6": ">=0.2.6"` (one line). |
| `pnpm-lock.yaml` | `tmp@0.2.4` → `tmp@0.2.6` resolved; `@scalar/nestjs-api-reference@1.1.16 → 1.1.19` picked up as a natural transitive resolution during `pnpm install` (no semver-major, both are within the existing `^1.1.14` range and match Renovate's "patch + auto-merge" policy). |

## Risk

Patch bump (0.2.4 → 0.2.6) on a tiny library with a stable public API since v0.2.0. The release notes for 0.2.5 and 0.2.6 are sanitization-only fixes — no API surface change. Risk of regression in nx / lhci consumers: negligible.

## Test plan

- [x] `pnpm audit --audit-level=moderate` returns `No known vulnerabilities found` (exit 0).
- [x] `pnpm why tmp` shows `tmp@0.2.6` as the single resolved version (no duplicate).
- [ ] CI green on Gitea (`check` + `scan` jobs).
- [ ] Once merged, rebase `ci/gitlab-pipeline-phase2` on top and retry the GitLab `audit` job — expected green.

## Related

- [GHSA-ph9p-34f9-6g65](https://github.com/advisories/GHSA-ph9p-34f9-6g65) — the advisory.
- Unblocks [ADR-0028](../docs/decisions/0028-migrate-cicd-and-git-hosting-to-gitlab.md) Phase 2 pipeline parity validation (see `ci/gitlab-pipeline-phase2` branch).

---------

Co-authored-by: Julien Gautier <julien.gautier@apf.asso.fr>
Reviewed-on: #241
This commit was merged in pull request #241.
This commit is contained in:
2026-05-27 17:53:39 +02:00
parent a848e0fabd
commit c24795f1d6
+149
View File
@@ -0,0 +1,149 @@
# Per ADR-0028 (CI/CD migration from Gitea Actions to GitLab CE).
# Thin YAML — orchestration lives in package.json scripts (ci:check,
# ci:catalogue-drift, ci:audit, ci:commits, ci:perf, ci:gzip-budgets)
# and Nx targets. Any change to gate behaviour belongs in those
# scripts, not in this file — see ADR-0015 §"Thin pipeline YAML…"
# (principle preserved by ADR-0028 at the migration boundary).
#
# Phase 2 of ADR-0028: ships alongside .gitea/workflows/ci.yml. Both
# pipelines must remain in parity until cutover (Phase 3), at which
# point the Gitea workflow gets deleted.
include:
# GitLab built-in scanners — replace the manual Trivy + gitleaks
# install in .gitea/workflows/ci.yml. Both add their own jobs to
# the pipeline; their default stage is `test`, which is mapped in
# `stages:` below.
- template: Jobs/SAST.gitlab-ci.yml
- template: Jobs/Secret-Detection.gitlab-ci.yml
stages:
- check
- test
- commits
- perf
- a11y
workflow:
rules:
# MR pipelines — covers GitLab MR review flow once it lands.
- if: $CI_PIPELINE_SOURCE == "merge_request_event"
# Direct pushes — covers post-merge runs on main today, and Phase 2
# parity testing on feature branches mirrored from Gitea. To be
# tightened to `$CI_COMMIT_BRANCH == "main"` at the Phase 3 cutover
# once GitLab is the source of truth.
- if: $CI_PIPELINE_SOURCE == "push"
# Shared shape for jobs that need Node + pnpm. Hidden (`.` prefix) so
# GitLab does not try to execute it. Inheriting jobs `extends: .node-job`.
.node-job:
image: node:24-bookworm
cache:
key:
files:
- pnpm-lock.yaml
paths:
- .pnpm-store/
before_script:
- corepack enable
- corepack prepare pnpm@10.33.4 --activate
- pnpm config set store-dir "$CI_PROJECT_DIR/.pnpm-store"
check:
extends: .node-job
stage: check
# `nx affected` needs a base SHA to diff against. Mirrors the
# equivalent step in .gitea/workflows/ci.yml — same logic, GitLab
# variable names. Replaces nrwl/nx-set-shas (GitHub-only).
script:
- |
if [ "$CI_PIPELINE_SOURCE" = "merge_request_event" ]; then
git fetch origin "$CI_MERGE_REQUEST_TARGET_BRANCH_NAME" --depth=100
export NX_BASE=$(git merge-base HEAD "origin/$CI_MERGE_REQUEST_TARGET_BRANCH_NAME")
else
export NX_BASE="HEAD~1"
fi
export NX_HEAD="HEAD"
- pnpm install --frozen-lockfile
- pnpm ci:check
# Catalogue-vs-code drift gate per ADR-0025 §"Confirmation". The
# script's own unit tests run first — if they fail the gate itself
# is broken and the actual catalogue check would be noise.
- pnpm ci:catalogue-drift:test
- pnpm ci:catalogue-drift
audit:
extends: .node-job
stage: test
# npm-advisory check against pnpm-lock.yaml. Trivy's dependency-vuln
# role from the Gitea workflow is now covered by the SAST include
# (which includes Dependency Scanning analyzers); `pnpm audit` stays
# in-pipeline as defense in depth against the npm advisory DB
# specifically.
script:
- pnpm install --frozen-lockfile
- pnpm ci:audit
commits:
extends: .node-job
stage: commits
# MRs opened by Renovate (apf-portal-bot) carry commit messages from
# a vetted Conventional-Commits template — running commitlint on
# them is tautological. Per ADR-0017 amendment.
rules:
- if: $CI_PIPELINE_SOURCE == "merge_request_event" && $GITLAB_USER_LOGIN != "apf-portal-bot"
variables:
# Default `GIT_DEPTH: 20` is too shallow to diff against
# origin/main on long-running branches. Full clone so commitlint
# can walk all the way back.
GIT_DEPTH: 0
script:
- git fetch origin main
- pnpm install --frozen-lockfile
- COMMIT_LINT_FROM=origin/main pnpm ci:commits
perf:
extends: .node-job
stage: perf
# Lighthouse CI drives a real Chrome instance. Playwright's image is
# the canonical "Node + browsers" CI image and is already on our
# roadmap for the ADR-0016 axe-core e2e — single image covers both
# uses once the a11y suite lands.
image: mcr.microsoft.com/playwright:v1.55.0-jammy
# Skip on Renovate MRs (same rationale as commits + the perf signal
# on a dep bump is essentially zero). Push events on `main` still
# run perf — we catch regressions immediately post-merge, not
# pre-merge. Per ADR-0017 amendment.
rules:
- if: $CI_PIPELINE_SOURCE == "push" && $CI_COMMIT_BRANCH == "main"
- if: $CI_PIPELINE_SOURCE == "merge_request_event" && $GITLAB_USER_LOGIN != "apf-portal-bot"
script:
- pnpm install --frozen-lockfile
- pnpm ci:perf
artifacts:
when: always
paths:
- .lighthouseci/
expire_in: 30 days
a11y:
extends: .node-job
stage: a11y
# Placeholder until the e2e a11y suite (axe-core via Playwright, per
# ADR-0016) is wired with the first real screens. The job exists so
# branch protection can require it from day one — currently no-ops
# with a clear message.
script:
- echo "a11y gate placeholder - axe-core via Playwright wires up with the first real screens (ADR-0016)."
# Place SAST + Secret Detection in the `test` stage to mirror the
# Gitea workflow's grouping (where Trivy + gitleaks + audit all
# lived in the `scan` job). Override does not change behaviour —
# only stage placement. The included templates set sensible defaults
# (run on default branch + MRs, full repo on default branch / diff
# on MRs); revisited in Phase 3 if pipeline duration becomes a pain.
sast:
stage: test
secret_detection:
stage: test