validate.yml

name: Content Validation

on:
  push:
    branches: [main]
    paths:
      - 'llm-top10/**'
      - 'agentic-top10/**'
      - 'dsgai-2026/**'
      - 'shared/**'
      - 'data/**'
      - 'CROSSREF.md'
      - 'README.md'
      - 'scripts/validate.js'
  pull_request:
    branches: [main]
    paths:
      - 'llm-top10/**'
      - 'agentic-top10/**'
      - 'dsgai-2026/**'
      - 'shared/**'
      - 'data/**'
      - 'CROSSREF.md'
      - 'README.md'
      - 'scripts/validate.js'

jobs:
  validate:
    name: Structural validation
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Use Node.js
        uses: actions/setup-node@v4
        with:
          node-version: '20'

      - name: Run content validator
        run: node scripts/validate.js

      - name: Validate JSON schema
        run: |
          node -e "
          const fs = require('fs');
          const schema = JSON.parse(fs.readFileSync('data/schema.json', 'utf8'));
          console.log('schema.json is valid JSON');
          console.log('Title:', schema.title);
          console.log('Required fields:', schema.required.join(', '));
          "

  mapping-counts:
    name: Count consistency
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Check mapping file counts match README badge
        run: |
          LLM=$(ls llm-top10/*.md | wc -l)
          AGT=$(ls agentic-top10/*.md | wc -l)
          DSG=$(ls dsgai-2026/*.md | wc -l)
          TOTAL=$((LLM + AGT + DSG))
          echo "LLM: $LLM  Agentic: $AGT  DSGAI: $DSG  Total: $TOTAL"

          BADGE=$(grep -o 'mapping%20files-[0-9]*-brightgreen' README.md | grep -o '[0-9]*')
          echo "README badge: $BADGE"

          if [ "$TOTAL" != "$BADGE" ]; then
            echo "::error::Count mismatch — $TOTAL files on disk but README badge says $BADGE"
            exit 1
          fi
          echo "Counts consistent: $TOTAL"