ci.yml

name: CI — Lint & Validate

on:
  pull_request:
    branches: [main]
  push:
    branches: [main]
    paths:
      - 'scripts/**'
      - 'config.yml'
      - 'requirements.txt'
      - '.github/workflows/**'

permissions: read-all
jobs:
  lint-and-validate:
    name: Python Lint & Syntax Check
    runs-on: ubuntu-latest
    permissions:
      contents: read

    steps:
      - name: Checkout repository
        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4

      - name: Set up Python
        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
        with:
          python-version: '3.11'
          cache: 'pip'

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -r requirements.txt
          pip install flake8

      - name: Syntax check — update_jobs.py
        run: python -m py_compile scripts/update_jobs.py

      - name: Syntax check — generate_companies.py
        run: python -m py_compile scripts/generate_companies.py

      - name: Syntax check — test_config.py
        run: python -m py_compile test_config.py

      - name: Syntax check — fix_nan_only.py
        run: python -m py_compile scripts/fix_nan_only.py

      - name: Flake8 lint (errors only, E9/F-class)
        run: |
          # Stop on syntax errors and undefined names only.
          # Style warnings (E501 long lines, W503, etc.) are informational.
          flake8 scripts/ --count --select=E9,F63,F7,F82 --show-source --statistics

      - name: Validate config.yml structure
        run: |
          python - <<'EOF'
          import yaml, sys
          with open("config.yml") as f:
              config = yaml.safe_load(f)
          required_keys = ["filtering", "readme"]
          for key in required_keys:
              if key not in config:
                  print(f"ERROR: Missing required top-level key in config.yml: '{key}'")
                  sys.exit(1)
          print(f"✅ config.yml is valid YAML with all required keys.")
          greenhouse = config.get("greenhouse", {}).get("companies", [])
          lever = config.get("lever", {}).get("companies", [])
          print(f"   Greenhouse companies: {len(greenhouse)}")
          print(f"   Lever companies: {len(lever)}")
          EOF

      - name: Run unit tests
        run: python test_config.py

      - name: Check for duplicate API URLs in config.yml
        run: |
          python - <<'EOF'
          import yaml, sys
          with open("config.yml") as f:
              config = yaml.safe_load(f)
          urls = []
          for section in ['greenhouse', 'lever', 'workday']:
              companies = config.get('apis', config).get(section, {}).get('companies', [])
              for c in companies:
                  url = c.get('url', c.get('workday_url', ''))
                  if url:
                      urls.append((section, c.get('name', ''), url))
          seen = {}
          duplicates = []
          for section, name, url in urls:
              if url in seen:
                  duplicates.append(f"  DUPLICATE: {url} ({name} in {section}) already defined as {seen[url]}")
              else:
                  seen[url] = f"{name} in {section}"
          if duplicates:
              print("❌ Duplicate API URLs found in config.yml:")
              for d in duplicates:
                  print(d)
              sys.exit(1)
          print(f"✅ No duplicate URLs found across {len(urls)} entries.")
          EOF