jekyll-docker.yml

name: Validate HTML & Assets

on:
  push:
    branches: [ "master" ]
  pull_request:
    branches: [ "master" ]

jobs:
  validate:
    runs-on: ubuntu-latest

    steps:
    - uses: actions/checkout@v4

    - name: Install validation tools
      run: sudo apt-get update -qq && sudo apt-get install -y -qq tidy libxml2-utils

    - name: Validate HTML syntax
      run: tidy -q -e index.html

    - name: Check for broken internal links
      run: |
        # Extract href/src values and verify local files exist
        exit_code=0
        for file in $(grep -oP '(?:href|src)="(?!https?://|//|#|mailto:)([^"]+)"' index.html | grep -oP '"[^"]+"' | tr -d '"'); do
          if [ ! -f "$file" ] && [ ! -d "$file" ]; then
            echo "ERROR: Missing local file referenced in index.html: $file"
            exit_code=1
          fi
        done
        exit $exit_code

    - name: Validate sitemap.xml well-formedness
      run: xmllint --noout sitemap.xml

    - name: Validate JSON-LD structured data syntax
      run: |
        # Extract JSON-LD blocks and validate JSON syntax
        python3 -c "
        import json, re, sys
        with open('index.html') as f:
            html = f.read()
        blocks = re.findall(r'<script type=\"application/ld\+json\">(.*?)</script>', html, re.DOTALL)
        if not blocks:
            print('WARNING: No JSON-LD blocks found')
            sys.exit(0)
        for i, block in enumerate(blocks):
            try:
                data = json.loads(block)
                print(f'JSON-LD block {i+1}: valid ({data.get(\"@type\", \"unknown type\")})')
            except json.JSONDecodeError as e:
                print(f'ERROR: JSON-LD block {i+1} is invalid: {e}')
                sys.exit(1)
        "

    - name: Validate robots.txt references
      run: |
        # Check that sitemap URL in robots.txt points to an existing file
        sitemap_path=$(grep -oP 'Sitemap:\s*https?://[^/]+/\K.*' robots.txt || true)
        if [ -n "$sitemap_path" ] && [ ! -f "$sitemap_path" ]; then
          echo "ERROR: robots.txt references missing sitemap: $sitemap_path"
          exit 1
        fi
        echo "robots.txt is valid"

    - name: Check no TODO/FIXME markers in HTML
      run: |
        if grep -inE '(TODO|FIXME|HACK|XXX)' index.html; then
          echo "WARNING: Found TODO/FIXME markers in index.html"
        else
          echo "No TODO/FIXME markers found"
        fi