content-validation.yaml

name: Content Validation

on:
  push:
    branches: [main]
    paths:
      - 'docs/**'
      - 'scripts/**'
  pull_request:
    branches: [main]
    paths:
      - 'docs/**'
      - 'scripts/**'
  workflow_dispatch:

jobs:
  validate:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout docs repo
        uses: actions/checkout@v4

      - name: Checkout PACK-personal repo
        uses: actions/checkout@v4
        with:
          repository: aisystant/PACK-personal
          path: PACK-personal
          token: ${{ secrets.PACK_READ_TOKEN || secrets.GITHUB_TOKEN }}

      - name: Setup Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Install dependencies
        run: |
          pip install pyyaml python-frontmatter

      - name: Markdown lint (validate.py)
        run: |
          python3 scripts/convert_word/validate.py docs/ru/personal-design/

      - name: Ontology drift check
        run: |
          python3 scripts/sync-guide-to-ontology.py --check \
            --ontology PACK-personal/ontology.md \
            --guides docs/ru/personal-design/
        env:
          PYTHONPATH: ${{ github.workspace }}

      - name: Broken cross-repo link detection
        run: |
          echo "Checking for broken ../../../PACK-personal/ontology.md# links..."
          python3 << 'EOF'
          import os, re
          ontology_path = "PACK-personal/ontology.md"
          if not os.path.exists(ontology_path):
              print("⚠️ SKIP: PACK-personal/ontology.md not found — link check skipped")
              exit(0)
          # Build set of anchors from ontology headings
          anchors = set()
          with open(ontology_path, "r", encoding="utf-8") as fh:
              for line in fh:
                  m = re.match(r"^(#{1,6})\s+(.+)$", line)
                  if m:
                      heading = m.group(2).strip()
                      # GitHub-style anchor: lowercase, spaces->-, remove punctuation
                      anchor = heading.lower().replace(" ", "-")
                      anchor = re.sub(r"[^\w\-]", "", anchor)
                      anchors.add(anchor)
          errors = []
          for root, dirs, files in os.walk("docs/ru/personal-design"):
              for f in files:
                  if not f.endswith(".md"):
                      continue
                  path = os.path.join(root, f)
                  with open(path, "r", encoding="utf-8") as fh:
                      content = fh.read()
                  for match in re.finditer(r'\.\./\.\./\.\./PACK-personal/ontology\.md#([^ )\]]+)', content):
                      link_anchor = match.group(1)
                      # Normalize anchor
                      normalized = link_anchor.lower().replace("%20", "-").replace(" ", "-")
                      normalized = re.sub(r"[^\w\-]", "", normalized)
                      if normalized not in anchors:
                          errors.append(f"{path}: broken anchor '#{link_anchor}'")
          if errors:
              print("❌ FAIL: Broken cross-repo links found:")
              for e in errors:
                  print(f"  - {e}")
              exit(1)
          else:
              print("✅ PASS: No broken cross-repo links found")
          EOF

      - name: Frontmeta completeness check
        run: |
          python3 << 'EOF'
          import os, re, yaml
          errors = []
          required_fields = [
              "title", "order", "introduces", "uses", "prerequisites",
              "can_do", "cp_check", "bh_check", "format_version",
              "pack_refs", "time_reading", "time_practice",
              "word_count_target", "status"
          ]
          for root, dirs, files in os.walk("docs/ru/personal-design"):
              for f in files:
                  if not f.endswith(".md"):
                      continue
                  # Пропускаем index.md и файлы без subsection-семантики
                  if f == "index.md":
                      continue
                  path = os.path.join(root, f)
                  with open(path, "r", encoding="utf-8") as fh:
                      content = fh.read()
                  m = re.match(r"^---\s*\n(.*?)\n---\s*\n", content, re.DOTALL)
                  if not m:
                      # Файлы без frontmatter — не подразделы, пропускаем
                      continue
                  try:
                      fm = yaml.safe_load(m.group(1)) or {}
                  except Exception as e:
                      errors.append(f"{path}: invalid YAML: {e}")
                      continue
                  # Проверяем только файлы с 'introduces' — это подразделы v4
                  if "introduces" not in fm:
                      continue
                  missing = [field for field in required_fields if field not in fm]
                  if missing:
                      errors.append(f"{path}: missing fields: {', '.join(missing)}")
          if errors:
              print("❌ Frontmatter completeness FAIL:")
              for e in errors:
                  print(f"  - {e}")
              exit(1)
          else:
              print("✅ Frontmatter completeness PASS")
          EOF

      - name: Didactic language check (UB-1)
        run: |
          echo "Checking for didactic language (UB-1)..."
          # Проверяем docs/ru/personal-design/ на запрещённые слова
          # «шаг» и «step» — легитимные слова в блоке «Минимальный шаг», исключены.
          # «урок», «внедрить», «за N дней», «lesson», «implement» — признаки дидактики.
          PATTERNS="урок|внедрить|за [0-9]+ дней|implement|lesson"
          if grep -riE "$PATTERNS" docs/ru/personal-design/; then
            echo "⚠️ WARN: Potential didactic language found (see above). Review manually."
          else
            echo "✅ PASS: No didactic language detected"
          fi