skos-monitor.yml

name: PRIMA SKOS Monitor

on:
  schedule:
    # Run weekly on Mondays at 10:00 UTC
    - cron: '0 10 * * 1'
  workflow_dispatch:
    # Allow manual triggering

jobs:
  monitor:
    runs-on: ubuntu-latest
    permissions:
      issues: write
      contents: write

    steps:
      - name: Checkout repository
        uses: actions/checkout@v6

      - name: Set up Python
        uses: actions/setup-python@v6
        with:
          python-version: '3.13'

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install requests python-dateutil rdflib PyGithub

      - name: Download latest SKOS
        id: download
        run: |
          python -m monitor_utils.prima_skos_monitor.cli download \
            --base-url "${{ env.SKOSMOS_BASE_URL || 'https://matwerk.datamanager.kit.edu/skosmos/prima/en/' }}" \
            --vocab-id "${{ env.SKOSMOS_VOCAB_ID || 'prima' }}" \
            --format rdf \
            --save-snapshot \
            --snapshot-dir monitor_utils/skos_snapshots \
            --output monitor_utils/skos_snapshots/latest_download.rdf
        env:
          SKOSMOS_BASE_URL: ${{ vars.SKOSMOS_BASE_URL || 'https://matwerk.datamanager.kit.edu/skosmos/prima/en/' }}
          SKOSMOS_VOCAB_ID: ${{ vars.SKOSMOS_VOCAB_ID || 'prima' }}

      - name: Find latest baseline snapshot
        id: baseline
        run: |
          if [ -f "monitor_utils/skos_snapshots/latest.rdf" ]; then
            BASELINE=$(readlink -f monitor_utils/skos_snapshots/latest.rdf || echo "monitor_utils/skos_snapshots/latest.rdf")
            echo "baseline_path=$BASELINE" >> $GITHUB_OUTPUT
            echo "has_baseline=true" >> $GITHUB_OUTPUT
          else
            # Use the downloaded file as both baseline and new (first run)
            echo "baseline_path=monitor_utils/skos_snapshots/latest_download.rdf" >> $GITHUB_OUTPUT
            echo "has_baseline=false" >> $GITHUB_OUTPUT
          fi

      - name: Compare versions
        id: compare
        run: |
          python -m monitor_utils.prima_skos_monitor.cli compare \
            --baseline "${{ steps.baseline.outputs.baseline_path }}" \
            --new monitor_utils/skos_snapshots/latest_download.rdf \
            --format rdf \
            --output change_report.md || true
        continue-on-error: true

      - name: Check for changes
        id: changes
        run: |
          if [ -f "change_report.md" ]; then
            # Check if report indicates changes (look for "No changes detected" or change counts)
            if grep -q "No changes detected" change_report.md; then
              echo "has_changes=false" >> $GITHUB_OUTPUT
            elif grep -qE "(New Concepts|Deleted Concepts|Modified Concepts)" change_report.md; then
              echo "has_changes=true" >> $GITHUB_OUTPUT
            else
              echo "has_changes=false" >> $GITHUB_OUTPUT
            fi
          else
            echo "has_changes=false" >> $GITHUB_OUTPUT
          fi

      - name: Create GitHub Issue
        if: steps.changes.outputs.has_changes == 'true'
        uses: actions/github-script@v7
        with:
          github-token: ${{ secrets.GITHUB_TOKEN }}
          script: |
            const fs = require('fs');
            const report = fs.readFileSync('change_report.md', 'utf8');
            
            const title = '[SKOS Monitor] Changes detected in PRIMA SKOS vocabulary';
            
            const body = `## Automated SKOS Change Detection
            
            This issue was automatically created by the PRIMA SKOS Monitor workflow.
            
            ${report}
            
            ---
            
            **Workflow Run:** ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
            **Triggered by:** ${{ github.event_name }}
            `;
            
            await github.rest.issues.create({
              owner: context.repo.owner,
              repo: context.repo.repo,
              title: title,
              body: body,
              labels: ['automated', 'skos-monitor']
            });

      - name: Save new snapshot
        if: steps.changes.outputs.has_changes == 'true' || steps.baseline.outputs.has_baseline == 'false'
        run: |
          # The snapshot was already saved by the download step
          # Just ensure latest.rdf is updated
          cd monitor_utils/skos_snapshots
          LATEST_SNAPSHOT=$(ls -t *.rdf | grep -v latest | head -1)
          if [ -n "$LATEST_SNAPSHOT" ]; then
            ln -sf "$LATEST_SNAPSHOT" latest.rdf
          fi

      - name: Commit snapshot and report
        if: steps.changes.outputs.has_changes == 'true' || steps.baseline.outputs.has_baseline == 'false'
        run: |
          git config --local user.email "action@github.com"
          git config --local user.name "GitHub Action"
          git add monitor_utils/skos_snapshots/ change_report.md
          git commit -m "chore: update SKOS snapshot and change report [skip ci]" || exit 0
          git push || exit 0

      - name: Summary
        run: |
          if [ "${{ steps.changes.outputs.has_changes }}" == "true" ]; then
            echo "## Changes Detected" >> $GITHUB_STEP_SUMMARY
            echo "A GitHub issue has been created with the change details." >> $GITHUB_STEP_SUMMARY
            echo "" >> $GITHUB_STEP_SUMMARY
            echo "See change_report.md for details." >> $GITHUB_STEP_SUMMARY
          else
            echo "## No Changes Detected" >> $GITHUB_STEP_SUMMARY
            echo "The PRIMA SKOS vocabulary has not changed since the last check." >> $GITHUB_STEP_SUMMARY
          fi