learningcircuit--local-deep-research/.github/workflows/playwright-webkit-tests.yml

name: Playwright WebKit Tests

on:
  workflow_call:      # Called by release pipeline
  workflow_dispatch:  # Manual trigger
  schedule:
    - cron: '0 2 * * *'  # Daily at 2 AM UTC

permissions: {}  # Minimal top-level for OSSF Scorecard Token-Permissions

jobs:
  # Shared setup: start server, register user, then run Desktop Safari tests
  desktop-safari:
    runs-on: ubuntu-latest
    permissions:
      contents: read

    steps:
    - name: Harden the runner (Audit all outbound calls)
      uses: step-security/harden-runner@9af89fc71515a100421586dfdb3dc9c984fbf411 # v2.19.4
      with:
        egress-policy: audit

    - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
      with:
        persist-credentials: false

    - name: Set up PDM
      uses: pdm-project/setup-pdm@973541a5febeafcfdadf8a51211435be6ecfd90f # v4.5
      with:
        python-version: '3.12'
        version: 2.26.2
        cache: true

    - name: Install Python dependencies
      run: |
        pdm install -d

    - name: Setup Node.js
      uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
      with:
        node-version: '24'
        cache: 'npm'
        cache-dependency-path: |
          package-lock.json
          tests/ui_tests/playwright/package-lock.json

    - name: Build Vite frontend assets
      run: |
        npm ci
        npm run build

    - name: Install Playwright dependencies (npm)
      working-directory: tests/ui_tests/playwright
      run: npm ci

    # ``playwright install --with-deps`` downloads ~150 MB of browser
    # binaries from the playwright CDN per attempt. Single CDN
    # hiccups have failed the whole release pipeline in the past
    # (e.g. transient 5xx mid-download); a single retry costs little
    # and removes the dominant transient-network failure mode for
    # this workflow. Two attempts only — three attempts of a 150 MB
    # download starts hiding real outages.
    - name: Install Playwright browsers
      uses: nick-fields/retry@ad984534de44a9489a53aefd81eb77f87c70dc60 # v4.0.0
      with:
        timeout_minutes: 10
        max_attempts: 2
        retry_on: error
        shell: bash
        # Install chromium for auth setup step + webkit for Safari tests
        command: |
          cd tests/ui_tests/playwright
          npx playwright install chromium webkit --with-deps

    - name: Setup test directories
      run: |
        mkdir -p data/encrypted_databases
        echo "Created data directories for tests"

    - name: Start LDR server
      env:
        CI: true
        TEST_ENV: true
        FLASK_ENV: testing
        LDR_DATA_DIR: ${{ github.workspace }}/data
        # The server MUST open test_admin's encrypted DB with the same KDF it
        # was created with. init_test_database.py creates it at 1000 iterations
        # (LDR_TEST_MODE relaxes the floor to allow it); without these two vars
        # here the requested 1000 is below the production floor, so it falls
        # back to the 256000 default, the server derives the wrong key, and
        # every login fails with a 401 — see #4558, which set these on the init
        # step but missed this server step.
        LDR_DB_CONFIG_KDF_ITERATIONS: "1000"
        LDR_TEST_MODE: "1"
        LDR_DISABLE_RATE_LIMITING: true
        SECRET_KEY: test-secret-key-for-ci
      run: |
        export PYTHONPATH="$PWD/src:$PYTHONPATH"
        echo "Starting server with LDR_DATA_DIR=$LDR_DATA_DIR"
        pdm run python -m local_deep_research.web.app &
        SERVER_PID=$!
        echo "SERVER_PID=$SERVER_PID" >> "$GITHUB_ENV"

        # Wait for server to be ready.
        # --connect-timeout/--max-time bound TCP and total request time so a
        # hung connection fails fast instead of eating the job timeout.
        for i in {1..30}; do
          if curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; then
            echo "Server is ready"
            break
          fi
          echo "Waiting for server... ($i/30)"
          sleep 2
        done

        # Verify server is running
        if ! curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; then
          echo "Server failed to start"
          exit 1
        fi

    - name: Initialize test database
      env:
        TEST_ENV: true
        LDR_DATA_DIR: ${{ github.workspace }}/data
        LDR_DB_CONFIG_KDF_ITERATIONS: "1000"
        # Without LDR_TEST_MODE the 1000 is clamped to the production
        # minimum (256000), making SQLCipher ~256x slower — see #4430.
        LDR_TEST_MODE: "1"
      run: |
        export PYTHONPATH="$PWD/src:$PYTHONPATH"
        pdm run python scripts/ci/init_test_database.py

    - name: Register CI test user
      working-directory: tests/ui_tests
      run: |
        npm ci
        node register_ci_user.js http://127.0.0.1:5000

    - name: Run Desktop Safari Tests
      working-directory: tests/ui_tests/playwright
      env:
        CI: true
        TEST_BASE_URL: http://127.0.0.1:5000
        TEST_USERNAME: test_admin
        TEST_PASSWORD: testpass123
      # Safari CI runs a critical subset for fast release gate:
      # - all-pages-mobile: overflow, touch targets, nav clearance across all pages
      # - auth-pages-mobile: login, register, change-password mobile layouts
      # - desktop-layout: sidebar positioning and breakpoint testing
      # - interactive-states: advanced options, search, help panels, mobile menus
      # - settings-subpages-mobile: settings tab navigation and form layouts
      # - mobile-ui-audit: comprehensive mobile UI quality checks
      # - embedding-settings-dropdown: regression for #3863 (model dropdown reset)
      # - settings-mobile-collapse: sections collapse by default on mobile (#4032),
      #   stay expanded on desktop, and search force-expands survivors
      # Full suite: run locally via `cd tests/ui_tests/playwright && npm test`
      # When adding Safari-critical tests, update this filter pattern.
      run: |
        npx playwright test \
          --project="Desktop Safari" \
          "all-pages-mobile|auth-pages-mobile|desktop-layout|interactive-states|settings-subpages-mobile|mobile-ui-audit|embedding-settings-dropdown|settings-mobile-collapse"

    - name: Upload Playwright Report
      if: always()
      uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
      with:
        name: playwright-report-desktop-safari
        path: |
          tests/ui_tests/playwright/playwright-report/
          tests/ui_tests/playwright/test-results/

    - name: Stop server
      if: always()
      run: |
        if [ -n "$SERVER_PID" ]; then
          kill "$SERVER_PID" || true
        fi
        pkill -f "python -m local_deep_research.web.app" || true

  # Mobile Safari tests run in parallel on a separate runner
  mobile-safari:
    runs-on: ubuntu-latest
    permissions:
      contents: read

    steps:
    - name: Harden the runner (Audit all outbound calls)
      uses: step-security/harden-runner@9af89fc71515a100421586dfdb3dc9c984fbf411 # v2.19.4
      with:
        egress-policy: audit

    - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
      with:
        persist-credentials: false

    - name: Set up PDM
      uses: pdm-project/setup-pdm@973541a5febeafcfdadf8a51211435be6ecfd90f # v4.5
      with:
        python-version: '3.12'
        version: 2.26.2
        cache: true

    - name: Install Python dependencies
      run: |
        pdm install -d

    - name: Setup Node.js
      uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
      with:
        node-version: '24'
        cache: 'npm'
        cache-dependency-path: |
          package-lock.json
          tests/ui_tests/playwright/package-lock.json

    - name: Build Vite frontend assets
      run: |
        npm ci
        npm run build

    - name: Install Playwright dependencies (npm)
      working-directory: tests/ui_tests/playwright
      run: npm ci

    # ``playwright install --with-deps`` downloads ~150 MB of browser
    # binaries from the playwright CDN per attempt. Single CDN
    # hiccups have failed the whole release pipeline in the past
    # (e.g. transient 5xx mid-download); a single retry costs little
    # and removes the dominant transient-network failure mode for
    # this workflow. Two attempts only — three attempts of a 150 MB
    # download starts hiding real outages.
    - name: Install Playwright browsers
      uses: nick-fields/retry@ad984534de44a9489a53aefd81eb77f87c70dc60 # v4.0.0
      with:
        timeout_minutes: 10
        max_attempts: 2
        retry_on: error
        shell: bash
        # Install chromium for auth setup step + webkit for Safari tests
        command: |
          cd tests/ui_tests/playwright
          npx playwright install chromium webkit --with-deps

    - name: Setup test directories
      run: |
        mkdir -p data/encrypted_databases
        echo "Created data directories for tests"

    - name: Start LDR server
      env:
        CI: true
        TEST_ENV: true
        FLASK_ENV: testing
        LDR_DATA_DIR: ${{ github.workspace }}/data
        # The server MUST open test_admin's encrypted DB with the same KDF it
        # was created with. init_test_database.py creates it at 1000 iterations
        # (LDR_TEST_MODE relaxes the floor to allow it); without these two vars
        # here the requested 1000 is below the production floor, so it falls
        # back to the 256000 default, the server derives the wrong key, and
        # every login fails with a 401 — see #4558, which set these on the init
        # step but missed this server step.
        LDR_DB_CONFIG_KDF_ITERATIONS: "1000"
        LDR_TEST_MODE: "1"
        LDR_DISABLE_RATE_LIMITING: true
        SECRET_KEY: test-secret-key-for-ci
      run: |
        export PYTHONPATH="$PWD/src:$PYTHONPATH"
        echo "Starting server with LDR_DATA_DIR=$LDR_DATA_DIR"
        pdm run python -m local_deep_research.web.app &
        SERVER_PID=$!
        echo "SERVER_PID=$SERVER_PID" >> "$GITHUB_ENV"

        # Wait for server to be ready.
        # --connect-timeout/--max-time bound TCP and total request time so a
        # hung connection fails fast instead of eating the job timeout.
        for i in {1..30}; do
          if curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; then
            echo "Server is ready"
            break
          fi
          echo "Waiting for server... ($i/30)"
          sleep 2
        done

        # Verify server is running
        if ! curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; then
          echo "Server failed to start"
          exit 1
        fi

    - name: Initialize test database
      env:
        TEST_ENV: true
        LDR_DATA_DIR: ${{ github.workspace }}/data
        LDR_DB_CONFIG_KDF_ITERATIONS: "1000"
        # Without LDR_TEST_MODE the 1000 is clamped to the production
        # minimum (256000), making SQLCipher ~256x slower — see #4430.
        LDR_TEST_MODE: "1"
      run: |
        export PYTHONPATH="$PWD/src:$PYTHONPATH"
        pdm run python scripts/ci/init_test_database.py

    - name: Register CI test user
      working-directory: tests/ui_tests
      run: |
        npm ci
        node register_ci_user.js http://127.0.0.1:5000

    - name: Run Mobile Safari Tests
      working-directory: tests/ui_tests/playwright
      env:
        CI: true
        TEST_BASE_URL: http://127.0.0.1:5000
        TEST_USERNAME: test_admin
        TEST_PASSWORD: testpass123
      # Safari CI runs a critical subset for fast release gate:
      # - all-pages-mobile: overflow, touch targets, nav clearance across all pages
      # - auth-pages-mobile: login, register, change-password mobile layouts
      # - desktop-layout: sidebar positioning and breakpoint testing
      # - interactive-states: advanced options, search, help panels, mobile menus
      # - settings-subpages-mobile: settings tab navigation and form layouts
      # - mobile-ui-audit: comprehensive mobile UI quality checks
      # - settings-mobile-collapse: sections collapse by default on mobile (#4032);
      #   this is the viewport where the >16384px overflow regression bites
      # Note: embedding-settings-dropdown is desktop-only (skips on isMobile),
      # so it intentionally only appears in the Desktop Safari filter above.
      # Full suite: run locally via `cd tests/ui_tests/playwright && npm test`
      # When adding Safari-critical tests, update this filter pattern.
      run: |
        npx playwright test \
          --project="Mobile Safari" \
          "all-pages-mobile|auth-pages-mobile|desktop-layout|interactive-states|settings-subpages-mobile|mobile-ui-audit|settings-mobile-collapse"

    - name: Upload Playwright Report
      if: always()
      uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
      with:
        name: playwright-report-mobile-safari
        path: |
          tests/ui_tests/playwright/playwright-report/
          tests/ui_tests/playwright/test-results/

    - name: Stop server
      if: always()
      run: |
        if [ -n "$SERVER_PID" ]; then
          kill "$SERVER_PID" || true
        fi
        pkill -f "python -m local_deep_research.web.app" || true