gstack/.github/workflows/free-tests.yml

name: Free Tests

# The free suite (~400 files: test/, browse/test/, make-pdf/test/, design/test/)
# had ZERO Linux CI coverage before this lane — only a curated Windows subset
# ran anywhere. This job runs the whole thing through the canonical runner
# (scripts/test-free-shards.ts): N concurrent shard processes (serial within
# each, plus a trailing serial tree-mutating shard) with strict-output
# classification per shard, so a truncated or summary-less run can never
# report green.
#
# Deliberately SECRETLESS: free tests make no API calls, so this lane gets no
# provider keys at all — least privilege, and fork PRs get real test signal
# here (the eval matrix skips fork PRs because repository secrets can't reach
# them). test/free-tests-workflow-wiring.test.ts fails CI if a secret sneaks in.
#
# This is a REQUIRED check from day one (branch protection lists it). If it's
# red, fix or quarantine-with-issue — don't make it advisory; an advisory lane
# is permanent false comfort.
#
# Sizing note (decision V3): single job first. If PR runs show it slower than
# the eval matrix wall, switch to a matrix of `--shards N --shard i` jobs
# (indices are stable, empty shards no-op).

on:
  pull_request:
    branches: [main]
  # Also on main pushes: two individually-green PRs can merge into a red
  # main; without this nothing runs the free suite on main until the next PR.
  push:
    branches: [main]
  workflow_dispatch:

# Keyed on the PR number, not head_ref: a bare branch name carries no fork
# prefix, so same-name branches from two forks would share one group and a
# push to fork B's PR would cancel fork A's in-flight REQUIRED check.
concurrency:
  group: free-tests-${{ github.event.pull_request.number || github.run_id }}
  cancel-in-progress: true

# Least privilege: this job executes PR-controlled code (install lifecycle
# scripts + the test suite), so the GITHUB_TOKEN gets read-only contents and
# the checkout doesn't persist it into .git/config.
permissions:
  contents: read

jobs:
  free-tests:
    runs-on: ubicloud-standard-8
    timeout-minutes: 20
    steps:
      - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
        with:
          persist-credentials: false

      - uses: oven-sh/setup-bun@v2
        with:
          bun-version: 1.3.13

      - uses: actions/cache@v6
        with:
          path: ~/.bun/install/cache
          key: linux-bun-${{ hashFiles('bun.lock') }}
          # A lockfile bump starts from the previous cache instead of cold.
          restore-keys: |
            linux-bun-

      - name: Install dependencies
        run: bun install --frozen-lockfile

      - uses: actions/cache@v6
        with:
          path: ~/.cache/ms-playwright
          key: linux-playwright-${{ hashFiles('bun.lock') }}
          restore-keys: |
            linux-playwright-

      # Cache restores browser binaries; install is still required for system
      # deps and is a fast no-op for already-present browsers.
      - name: Install Playwright Chromium
        run: npx playwright install --with-deps chromium

      # Headed-browser tests (handoff, extension sidepanel DOM) need a real
      # DISPLAY — first Linux run failed with Playwright's "launched a headed
      # browser without an XServer" banner. xvfb-run below provides it;
      # x11-utils ships xdpyinfo for display probing. poppler-utils ships
      # pdftotext/pdffonts/pdftoppm for the make-pdf e2e gates;
      # fonts-noto-color-emoji is the emoji-gate's render font (playwright
      # --with-deps usually installs it, but the gate must not depend on a
      # transitive package list). Fonts must land BEFORE the first browse
      # daemon launch — Chromium snapshots fontconfig at startup.
      - name: Install Xvfb + X11 utilities + gate tools
        run: sudo apt-get install -y --no-install-recommends xvfb x11-utils poppler-utils fonts-noto-color-emoji

      - name: Configure git identity (tests init temp repos)
        run: |
          git config --global user.email "free-tests-ci@gstack.test"
          git config --global user.name "Free Tests CI"
          git config --global init.defaultBranch main
          # Some tests run git against the checkout itself; CI checkouts can be
          # owned by a different uid than the runner user.
          git config --global --add safe.directory '*'

      - name: Generate host SKILL.md outputs (.agents, .factory)
        # Golden-file tests read generated host outputs that are gitignored.
        run: bun run gen:skill-docs --host all

      - name: Vendor xterm assets into the extension
        # extension/lib/xterm* are gitignored (vendored from npm at build
        # time). Without them the sidepanel's terminal script bails and the
        # sidepanel DOM tests time out waiting on init that never happens.
        run: bun run vendor:xterm

      - name: Build server-node bundle (loaded by browse cli imports)
        run: bash browse/scripts/build-node-server.sh

      # Narrowed gate build: the make-pdf e2e gates probe make-pdf/dist/pdf,
      # browse/dist/browse, and the diagram-render bundle, then self-skip when
      # absent — which made them silently skip on Linux for their whole life
      # (this lane never built binaries). Full `bun run build` compiles five
      # binaries and would add ~60-90s to the ONLY required check; the gates
      # need exactly these three artifacts.
      - name: Build gate binaries (make-pdf e2e gates)
        run: bun run build:gates

      # GSTACK_EXPECT_BINARIES=1 arms make-pdf/test/e2e/ci-prereqs.test.ts:
      # if a future edit drops the gate build (or poppler), the lane FAILS
      # instead of the gates silently self-skipping back to false green.
      - name: Run free suite
        run: xvfb-run -a bun run test:free
        env:
          GSTACK_EXPECT_BINARIES: "1"
          # WS1 flake telemetry: a single timing flake must not red the only
          # required lane — the runner's attribution-gated retry pass (cap 5,
          # truncation veto) re-runs failing files once, serially, and a
          # clean retry downgrades to a LOUD flaky-pass. Every flaky-pass is
          # appended to the ledger (single writer: the parent runner) and
          # uploaded below, so repeat offenders are an enumerable series —
          # recorded and ranked, never masked. Pinned by
          # test/free-tests-workflow-wiring.test.ts.
          GSTACK_FREE_RETRY_FLAKY: "1"
          GSTACK_FLAKE_LEDGER: ${{ runner.temp }}/flake-ledger.jsonl

      # Uploaded unconditionally (not just on failure): a flaky-pass run is
      # GREEN — that's the point — so its evidence must survive green runs.
      - name: Upload flake ledger
        if: always()
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
        with:
          name: flake-ledger
          path: ${{ runner.temp }}/flake-ledger.jsonl
          if-no-files-found: ignore
          retention-days: 90

      # The runner streams the full child output to per-run logs under the OS
      # tmpdir and prints only the quiet contract to the console. Without this
      # upload, a red required check names WHICH test failed but the why
      # (assertion detail, stack) dies with the runner — every diagnosis would
      # need a local re-run, which fork contributors can't do on this image.
      - name: Upload shard logs on failure
        if: failure()
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
        with:
          name: free-test-shard-logs
          path: /tmp/gstack-free-test-*.log
          if-no-files-found: ignore