update-data.yml

name: Auto-Update Stats

on:
  schedule:
    # Run every 6 hours to keep stats fresh
    - cron: '0 */6 * * *'
  workflow_dispatch: # Allow manual triggering
  push:
    branches:
      - main
    paths:
      - 'generate_html.py'
      - 'update_readme.py'
      - '.github/workflows/update-data.yml'

jobs:
  update-data:
    runs-on: ubuntu-latest

    permissions:
      contents: write

    steps:
      - name: Checkout repository
        uses: actions/checkout@v7
        with:
          token: ${{ secrets.GITHUB_TOKEN }}

      - name: Set up Python
        uses: actions/setup-python@v6
        with:
          python-version: '3.11'

      - name: Cache pip dependencies
        uses: actions/cache@v5
        with:
          path: ~/.cache/pip
          key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements.txt') }}
          restore-keys: |
            ${{ runner.os }}-pip-

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          # Install any dependencies if requirements.txt exists
          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi

      - name: Configure git
        run: |
          git config --local user.email "action@github.com"
          git config --local user.name "GitHub Action"

      - name: Verify stats.json accessibility
        run: |
          echo "Checking status.semcl.one/data/stats.json accessibility..."
          curl -f -s -o /tmp/stats.json https://status.semcl.one/data/stats.json
          echo "Last updated: $(jq -r '.last_updated' /tmp/stats.json)"
          echo "Packages found: $(jq '.packages | length' /tmp/stats.json)"
          rm /tmp/stats.json

      - name: Update README with latest component stats
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          echo "Running update_readme.py..."
          python3 update_readme.py

          if [ $? -eq 0 ]; then
            echo "✅ README update completed successfully"
          else
            echo "❌ README update failed"
            exit 1
          fi

      - name: Update website with latest component data
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          echo "Running generate_html.py..."
          python3 generate_html.py

          if [ $? -eq 0 ]; then
            echo "✅ Website update completed successfully"

            # Verify downloads stat was updated in HTML
            downloads=$(grep -B 1 'stat-label">Downloads' index.html | grep 'stat-number' | grep -oP '>\K[^<]+' || echo "not found")
            echo "Downloads stat in HTML: $downloads"

            if [ "$downloads" = "N/A" ] || [ "$downloads" = "n/a" ] || [ "$downloads" = "not found" ]; then
              echo "⚠️ Warning: Downloads stat is '$downloads' - data may not be loading correctly"
            fi
          else
            echo "❌ Website update failed"
            exit 1
          fi

      - name: Check for changes
        id: changes
        run: |
          if git diff --quiet; then
            echo "changes=false" >> $GITHUB_OUTPUT
            echo "No changes detected"
          else
            echo "changes=true" >> $GITHUB_OUTPUT
            echo "Changes detected:"
            git diff --name-only
          fi

      - name: Commit and push changes
        if: steps.changes.outputs.changes == 'true'
        run: |
          # Extract current downloads stat for commit message
          downloads=$(grep -B 1 'stat-label">Downloads' index.html | grep 'stat-number' | grep -oP '>\K[^<]+' || echo "unknown")

          git add README.md index.html
          git commit -m "chore: auto-update component stats [skip ci]

          - Downloads: $downloads (all-time, organic)
          - Updated at: $(date -u '+%Y-%m-%d %H:%M UTC')
          - Data source: status.semcl.one/data/stats.json"
          git push

      - name: Create summary
        run: |
          echo "## Auto-Update Stats Summary" >> $GITHUB_STEP_SUMMARY
          echo "" >> $GITHUB_STEP_SUMMARY

          # Show current stats
          downloads=$(grep -B 1 'stat-label">Downloads' index.html | grep 'stat-number' | grep -oP '>\K[^<]+' || echo "unknown")
          completion=$(grep -B 1 'stat-label">Complete' index.html | grep 'stat-number' | grep -oP '>\K[^<]+' || echo "unknown")
          components=$(grep -B 1 'stat-label">Components' index.html | grep 'stat-number' | grep -oP '>\K[^<]+' || echo "unknown")

          echo "### Current Stats" >> $GITHUB_STEP_SUMMARY
          echo "- 📥 **Downloads**: $downloads" >> $GITHUB_STEP_SUMMARY
          echo "- 📊 **Completion**: $completion" >> $GITHUB_STEP_SUMMARY
          echo "- 🎯 **Components**: $components" >> $GITHUB_STEP_SUMMARY
          echo "" >> $GITHUB_STEP_SUMMARY

          if [ "${{ steps.changes.outputs.changes }}" == "true" ]; then
            echo "### ✅ Changes Committed" >> $GITHUB_STEP_SUMMARY
            echo "" >> $GITHUB_STEP_SUMMARY
            echo "Updated files:" >> $GITHUB_STEP_SUMMARY
            git diff --name-only HEAD~1 HEAD | sed 's/^/- /' >> $GITHUB_STEP_SUMMARY
          else
            echo "### ℹ️ No Changes" >> $GITHUB_STEP_SUMMARY
            echo "" >> $GITHUB_STEP_SUMMARY
            echo "All component data is already up to date." >> $GITHUB_STEP_SUMMARY
          fi
          echo "" >> $GITHUB_STEP_SUMMARY
          echo "---" >> $GITHUB_STEP_SUMMARY
          echo "**Run completed at:** $(date -u '+%Y-%m-%d %H:%M:%S UTC')" >> $GITHUB_STEP_SUMMARY
          echo "**Data source:** https://status.semcl.one/data/stats.json" >> $GITHUB_STEP_SUMMARY