sanity-check.yml

name: PR Sanity Check

on:
  pull_request:
    types: [opened, synchronize, reopened]

permissions:
  pull-requests: write
  contents: read

jobs:
  sanity-check:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout code
        uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Fetch base branch
        run: git fetch origin ${{ github.base_ref }}

      - name: Block disallowed files by extension
        run: |
          echo "Checking for disallowed file extensions..."
          CHANGED_FILES=$(git diff --name-only origin/${{ github.base_ref }}...HEAD)

          # Block by extension
          BLOCKED=$(echo "$CHANGED_FILES" | grep -E '\.(exe|dll|out|ezdb|db|sqlite|sqlite3)$' || true)
          if [ -n "$BLOCKED" ]; then
            echo "::error::Disallowed file extensions detected: $BLOCKED"
            exit 1
          fi

          # Block IDE/OS junk folders
          JUNK=$(echo "$CHANGED_FILES" | grep -E '^(\.idea/|\.vs/|__pycache__/)' || true)
          if [ -n "$JUNK" ]; then
            echo "::error::IDE/OS junk detected: $JUNK"
            exit 1
          fi

          # Block OS junk files
          OS_JUNK=$(echo "$CHANGED_FILES" | grep -E '(Thumbs\.db|Desktop\.ini|\.DS_Store)$' || true)
          if [ -n "$OS_JUNK" ]; then
            echo "::error::OS junk files detected: $OS_JUNK"
            exit 1
          fi

          echo "No disallowed file extensions detected."

      - name: Block meta/config files from other ecosystems
        run: |
          echo "Checking for ecosystem files..."
          CHANGED_FILES=$(git diff --name-only origin/${{ github.base_ref }}...HEAD)

          # Block meta/config files from other language ecosystems
          SUSPICIOUS=$(echo "$CHANGED_FILES" | grep -E '^(package\.json|package-lock\.json|yarn\.lock|pnpm-lock\.yaml|bun\.lockb|bunfig\.toml|deno\.json|deno\.jsonc|tsconfig\.json|jsconfig\.json|requirements\.txt|Pipfile|Gemfile|Cargo\.toml|Cargo\.lock)$|node_modules/' || true)

          if [ -n "$SUSPICIOUS" ]; then
            echo "::error::Ecosystem config files detected: $SUSPICIOUS"
            exit 1
          fi
          echo "No suspicious ecosystem files detected."

      - name: Reject binary files except images
        run: |
          echo "Checking for binary files..."
          CHANGED_FILES=$(git diff --name-only origin/${{ github.base_ref }}...HEAD)

          for file in $CHANGED_FILES; do
            if [ -f "$file" ]; then
              MIME=$(file --mime-type -b "$file")

              # Allow common image formats
              if [[ "$MIME" =~ ^image/(jpeg|png|gif|svg\+xml|webp)$ ]]; then
                echo "Allowed image file: $file ($MIME)"
                continue
              fi

              # Allow shell scripts (they report as "executable" but are text files)
              if [[ "$file" =~ \.sh$ ]]; then
                echo "Allowed shell script: $file"
                continue
              fi

              # Allow WebAssembly binaries
              if [[ "$file" =~ \.wasm$ ]]; then
                echo "Allowed WASM file: $file"
                continue
              fi

              # Reject executables and archives (use -b to exclude filename from output)
              if file -b "$file" | grep -qE "executable|binary|archive|compressed"; then
                echo "::error::Binary/executable file detected: $file"
                exit 1
              fi

              # Also check charset=binary for other binary types (but not images)
              if file -b --mime "$file" | grep -q "charset=binary" && [[ ! "$MIME" =~ ^image/ ]]; then
                echo "::error::Binary file detected: $file ($MIME)"
                exit 1
              fi
            fi
          done
          echo "No disallowed binary files detected."

      - name: Close PR if checks fail
        if: failure()
        env:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          gh pr close ${{ github.event.pull_request.number }} \
            --repo ${{ github.repository }} \
            --comment "This PR has been automatically closed because it contains disallowed files (binaries, executables, archives, or meta/config files from other ecosystems). Please remove these files and open a new PR."