openhuman/.github/workflows/ci-full.yml

---
# CI Full — the slow lane. Runs EVERYTHING against the long-lived `release`
# branch: full frontend + Rust unit suites, Rust mock-backend E2E, Playwright
# web E2E, and the full desktop E2E matrix on Linux/macOS/Windows.
#
# Two-lane model:
#   - ci-lite.yml (quick): pushes to main + PRs to main/release — quality
#     checks + unit tests scoped to the changed files only.
#   - ci-full.yml (this file, slow): PRs targeting `release` (fix PRs opened
#     while stabilising a cut — the "CI Full Gate" check gates their merge)
#     and pushes to `release` (the maintainer-dispatched promotion merge from
#     promote-main-to-release.yml, which has no PR, plus each fix-PR merge
#     commit). Once the gate is green on release HEAD a release is cut via
#     release-staging.yml / release-production.yml.
#
# The release workflows' version-bump commits carry [skip ci] so cutting a
# build does not re-trigger this suite on an already-validated tree.
name: CI Full

on:
  push:
    branches: [release]
  pull_request:
    branches: [release]
  workflow_dispatch: {}

permissions:
  # `actions: read` lets scripts/ci-cancel-aware.sh poll the run status so
  # cancelled builds inside container jobs stop themselves (docker exec
  # swallows the runner's signals).
  actions: read
  contents: read
  packages: read

env:
  # Consumed by scripts/ci-cancel-aware.sh's cancellation watchdog.
  GH_TOKEN: ${{ github.token }}
  NPM_CONFIG_STORE_DIR: .pnpm-store

# Keep only the newest run per PR (or per ref for release pushes); the gate
# result is per-commit, so cancelling the superseded run loses nothing.
concurrency:
  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
  cancel-in-progress: true

defaults:
  run:
    shell: bash

jobs:
  # Full unit suites: frontend Vitest, i18n coverage, Rust core tests, Rust
  # Tauri shell tests — the same reusable recipe the release workflows'
  # pretest gate uses.
  unit-tests:
    name: Full Unit Suites
    # No `secrets: inherit` — the called workflow uses no secrets; keep this
    # workflow's runs secret-free.
    uses: ./.github/workflows/test-reusable.yml
    with:
      run_unit: true
      run_rust_core: true
      run_rust_tauri: true

  rust-e2e:
    name: Rust E2E (mock backend)
    runs-on: ubuntu-22.04
    container:
      image: ghcr.io/tinyhumansai/openhuman_ci:latest
    timeout-minutes: 30
    env:
      CARGO_INCREMENTAL: "0"
      # Trim DWARF debug info so test-binary linking doesn't exhaust runner disk
      # (SIGBUS). Tests don't need full debug info; backtraces keep file/line.
      CARGO_PROFILE_DEV_DEBUG: line-tables-only
    steps:
      - name: Free disk space
        run: |
          # Reclaim space on the build partition before compiling/linking the
          # Rust test binaries (host toolcache is bind-mounted at /__t inside
          # the container).
          rm -rf /__t/* || true
          df -h /__w || true

      - name: Checkout code
        uses: actions/checkout@v7
        with:
          fetch-depth: 1
          persist-credentials: false
          submodules: recursive

      - name: Cache pnpm store
        id: pnpm-cache
        uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .pnpm-store
          key: pnpm-store-${{ runner.os }}-${{ hashFiles('pnpm-lock.yaml') }}
          restore-keys: |
            pnpm-store-${{ runner.os }}-

      - name: Cache Rust build artifacts
        uses: Swatinem/rust-cache@v2
        with:
          workspaces: . -> target
          cache-on-failure: true
          # Shared with ci-lite's former key so warm caches carry over between
          # the fast lane and this full lane on the same dependency tree.
          shared-key: pr-rust-core

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

      - name: Save pnpm store cache
        if: always() && steps.pnpm-cache.outputs.cache-hit != 'true'
        uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .pnpm-store
          key: ${{ steps.pnpm-cache.outputs.cache-primary-key }}

      - name: Ensure .env exists for tests
        run: |
          touch .env
          touch app/.env

      - name: Run Rust E2E suite
        # Wrapped so the cancellation watchdog covers the whole suite (mock
        # server + node harness), not just the cargo segments the inner
        # script wraps itself.
        run: bash scripts/ci-cancel-aware.sh pnpm test:rust:e2e
        env:
          RUST_BACKTRACE: "1"

  build-playwright-e2e-artifact:
    name: Build Playwright E2E Artifact
    runs-on: ubuntu-22.04
    container:
      image: ghcr.io/tinyhumansai/openhuman_ci:latest
    timeout-minutes: 30
    env:
      CARGO_INCREMENTAL: "0"
    steps:
      - name: Checkout code
        uses: actions/checkout@v7
        with:
          fetch-depth: 1
          persist-credentials: false
          submodules: recursive

      - name: Cache pnpm store
        id: pnpm-cache
        uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .pnpm-store
          key: pnpm-store-${{ runner.os }}-${{ hashFiles('pnpm-lock.yaml') }}
          restore-keys: |
            pnpm-store-${{ runner.os }}-

      - name: Cache Rust build artifacts
        uses: Swatinem/rust-cache@v2
        with:
          workspaces: . -> target
          cache-on-failure: true
          shared-key: pr-rust-core

      - name: Restore cached Playwright E2E artifact
        id: playwright-artifact-cache
        uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .ci/artifacts/e2e-playwright-linux.tar.gz
          key: e2e-playwright-linux-${{ hashFiles('src/**', 'Cargo.toml', 'Cargo.lock', 'rust-toolchain.toml', 'app/src/**', 'app/public/**', 'app/index.html', 'app/vite.config.*', 'app/tailwind.config.*', 'app/postcss.config.*', 'app/package.json', 'pnpm-lock.yaml', 'app/scripts/e2e-web-build.sh', 'app/scripts/build-parallel.mjs') }}

      - name: Install dependencies
        if: steps.playwright-artifact-cache.outputs.cache-hit != 'true'
        run: pnpm install --frozen-lockfile

      - name: Save pnpm store cache
        if: always() && steps.pnpm-cache.outputs.cache-hit != 'true'
        uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .pnpm-store
          key: ${{ steps.pnpm-cache.outputs.cache-primary-key }}

      - name: Ensure .env exists for E2E build
        if: steps.playwright-artifact-cache.outputs.cache-hit != 'true'
        run: |
          touch .env
          touch app/.env

      - name: Build and package Playwright E2E artifact
        if: steps.playwright-artifact-cache.outputs.cache-hit != 'true'
        run: |
          set -euo pipefail
          bash scripts/ci-cancel-aware.sh pnpm --filter openhuman-app test:e2e:web:build

          RUST_HOST_TRIPLE="$(rustc -vV | awk '/^host: / { print $2 }')"
          WEB_CORE_TARGET_DIR="target/e2e-web-${RUST_HOST_TRIPLE}"
          STAGE="$(mktemp -d)"
          mkdir -p "$STAGE/repo/app/dist-web"
          mkdir -p "$STAGE/repo/$WEB_CORE_TARGET_DIR/debug"
          mkdir -p .ci/artifacts

          cp -a app/dist-web/. "$STAGE/repo/app/dist-web/"
          cp -a "$WEB_CORE_TARGET_DIR/debug/openhuman-core" "$STAGE/repo/$WEB_CORE_TARGET_DIR/debug/"
          tar -czf .ci/artifacts/e2e-playwright-linux.tar.gz -C "$STAGE" repo
          ls -lh .ci/artifacts/e2e-playwright-linux.tar.gz

      - name: Verify build artifact exists
        run: |
          set -euo pipefail
          test -f .ci/artifacts/e2e-playwright-linux.tar.gz

      - name: Save Playwright E2E artifact cache
        if: always() && steps.playwright-artifact-cache.outputs.cache-hit != 'true' && hashFiles('.ci/artifacts/e2e-playwright-linux.tar.gz') != ''
        uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .ci/artifacts/e2e-playwright-linux.tar.gz
          key: ${{ steps.playwright-artifact-cache.outputs.cache-primary-key }}

      - name: Upload Playwright E2E artifact
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
        with:
          name: e2e-playwright-linux-${{ github.run_id }}
          path: .ci/artifacts/e2e-playwright-linux.tar.gz
          retention-days: 1
          if-no-files-found: error

  playwright-e2e:
    name: E2E (Playwright / web lane)
    needs: [build-playwright-e2e-artifact]
    runs-on: ubuntu-22.04
    container:
      image: ghcr.io/tinyhumansai/openhuman_ci:latest
    timeout-minutes: 90
    # TODO(ci-flaky, #3615): several specs consistently hit toBeVisible
    # timeouts under CI resource contention. Run the lane but don't block
    # release merges until infra stabilises.
    continue-on-error: true
    steps:
      - name: Checkout code
        uses: actions/checkout@v7
        with:
          fetch-depth: 1
          persist-credentials: false
          submodules: recursive

      - name: Cache pnpm store
        id: pnpm-cache
        uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .pnpm-store
          key: pnpm-store-${{ runner.os }}-${{ hashFiles('pnpm-lock.yaml') }}
          restore-keys: |
            pnpm-store-${{ runner.os }}-

      - name: Install JS dependencies (test harness)
        run: pnpm install --frozen-lockfile

      - name: Save pnpm store cache
        if: always() && steps.pnpm-cache.outputs.cache-hit != 'true'
        uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
        with:
          path: .pnpm-store
          key: ${{ steps.pnpm-cache.outputs.cache-primary-key }}

      - name: Download Playwright E2E artifact
        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
        with:
          name: e2e-playwright-linux-${{ github.run_id }}
          path: .

      - name: Restore Playwright E2E artifact
        run: |
          set -euo pipefail
          tar -xzf e2e-playwright-linux.tar.gz
          cp -a repo/app/dist-web app/
          mkdir -p target
          cp -a repo/target/. target/
          rm -rf repo e2e-playwright-linux.tar.gz
          chmod +x target/e2e-web-*/debug/openhuman-core

      - name: Install Playwright Chromium headless shell
        run: pnpm --filter openhuman-app exec playwright install chromium-headless-shell

      - name: Run Playwright web E2E suite
        env:
          OPENHUMAN_WORKSPACE: ${{ runner.temp }}/openhuman-playwright-workspace
        run: |
          mkdir -p "$OPENHUMAN_WORKSPACE"
          bash scripts/ci-cancel-aware.sh bash app/scripts/e2e-web-session.sh

      - name: Pack Playwright E2E failure artifacts
        if: failure()
        run: |
          mkdir -p .ci/artifacts
          tar -czf .ci/artifacts/openhuman-playwright-failure-logs.tar.gz \
            -C "$OPENHUMAN_WORKSPACE" .
        env:
          OPENHUMAN_WORKSPACE: ${{ runner.temp }}/openhuman-playwright-workspace

      - name: Upload Playwright E2E failure artifacts
        if: failure()
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
        with:
          name: e2e-playwright-failure-logs-${{ github.run_id }}
          path: .ci/artifacts/openhuman-playwright-failure-logs.tar.gz
          retention-days: 7
          if-no-files-found: ignore

  # Full desktop E2E suite on all three shipped OSes (build-once-then-fanout
  # shards — see e2e-reusable.yml).
  e2e-desktop:
    name: Desktop E2E (full suite, 3 OS)
    # No `secrets: inherit` — the called workflow uses no secrets; keep this
    # workflow's runs secret-free.
    uses: ./.github/workflows/e2e-reusable.yml
    with:
      run_linux: true
      run_macos: true
      run_windows: true
      full: true

  ci-full-gate:
    name: CI Full Gate
    needs:
      - unit-tests
      - rust-e2e
      - build-playwright-e2e-artifact
      # `playwright-e2e` is intentionally EXCLUDED from the gate. It carries
      # `continue-on-error: true` (flaky specs, TODO #3615), which makes the
      # job report `success` to `needs` even when its steps fail — so listing
      # it here would be a can-never-fail check that only reads as coverage.
      # Rather than leave it decorative, we drop it from the gate until the
      # specs are stabilised and the flag removed (#3615); the lane still runs
      # for signal, and `build-playwright-e2e-artifact` (which CAN fail) stays
      # gated so a broken web build is still caught.
      - e2e-desktop
    if: always()
    runs-on: ubuntu-latest
    timeout-minutes: 5
    steps:
      - name: Require all CI Full jobs to pass
        run: |
          set -euo pipefail
          # NOTE: "Playwright E2E" is deliberately absent — see the needs:
          # comment above (#3615). Re-add it here once continue-on-error is
          # removed from the playwright-e2e job.
          declare -A results=(
            ["Full Unit Suites"]="${{ needs['unit-tests'].result }}"
            ["Rust E2E"]="${{ needs['rust-e2e'].result }}"
            ["Build Playwright E2E Artifact"]="${{ needs['build-playwright-e2e-artifact'].result }}"
            ["Desktop E2E"]="${{ needs['e2e-desktop'].result }}"
          )

          failed=0
          for name in "${!results[@]}"; do
            result="${results[$name]}"
            echo "${name}: ${result}"
            if [ "${result}" != "success" ] && [ "${result}" != "skipped" ]; then
              echo "::error::${name} did not pass (result=${result})."
              failed=1
            fi
          done

          if [ "${failed}" -ne 0 ]; then
            exit 1
          fi