update-data.yml

name: Update Data

on:
  # Run weekly on Sundays at 00:00 UTC
  schedule:
    - cron: '0 0 * * 0'

  # Allow manual trigger
  workflow_dispatch:

jobs:
  update-data:
    name: Update Timezone Data
    runs-on: ubuntu-latest
    permissions:
      contents: write
      pull-requests: write

    steps:
      - name: Checkout code
        uses: actions/checkout@v4
        with:
          token: ${{ secrets.GITHUB_TOKEN }}

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

      - name: Install dependencies
        run: npm ci

      - name: Download latest data
        run: |
          curl -L "https://raw.githubusercontent.com/dr5hn/countries-states-cities-database/refs/heads/master/json/countries%2Bstates%2Bcities.json.gz" \
            -o /tmp/countries-data.json.gz
          gunzip /tmp/countries-data.json.gz
          echo "Downloaded data size:"
          ls -lh /tmp/countries-data.json

      - name: Backup current data statistics
        run: |
          echo "=== Current Data Statistics ===" > /tmp/data-stats-before.txt
          if [ -f src/data/timezones.json ]; then
            echo "Timezones: $(grep -c '"zoneName"' src/data/timezones.json)" >> /tmp/data-stats-before.txt
            echo "Data files: $(find src/data -type f | wc -l)" >> /tmp/data-stats-before.txt
          else
            echo "No existing data found" >> /tmp/data-stats-before.txt
          fi
          cat /tmp/data-stats-before.txt

      - name: Generate timezone data
        run: node scripts/generate-data.cjs /tmp/countries-data.json

      - name: Collect new data statistics
        id: data-stats
        run: |
          echo "=== New Data Statistics ===" > /tmp/data-stats-after.txt
          echo "Timezones: $(grep -c '"zoneName"' src/data/timezones.json)" >> /tmp/data-stats-after.txt
          echo "Countries: $(grep -c '"countryCode"' src/data/timezones.json | head -1)" >> /tmp/data-stats-after.txt
          echo "Abbreviations: $(find src/data/abbreviations -type f -name '*.json' | wc -l)" >> /tmp/data-stats-after.txt
          cat /tmp/data-stats-after.txt

          # Export for PR description
          TIMEZONES=$(grep -c '"zoneName"' src/data/timezones.json)
          COUNTRIES=$(find src/data/countries -type f -name '*.json' | wc -l)
          ABBREVIATIONS=$(find src/data/abbreviations -type f -name '*.json' | wc -l)
          echo "timezones=$TIMEZONES" >> $GITHUB_OUTPUT
          echo "countries=$COUNTRIES" >> $GITHUB_OUTPUT
          echo "abbreviations=$ABBREVIATIONS" >> $GITHUB_OUTPUT

      - name: Build and test with new data
        run: |
          npm run build
          npm test

      - name: Configure git
        run: |
          git config --global user.name "github-actions[bot]"
          git config --global user.email "github-actions[bot]@users.noreply.github.com"

      - name: Check for changes
        id: check-changes
        run: |
          git add src/data
          if git diff --cached --quiet; then
            echo "changed=false" >> $GITHUB_OUTPUT
            echo "No data changes detected"
          else
            echo "changed=true" >> $GITHUB_OUTPUT
            echo "Data changes detected"
          fi

      - name: Create Pull Request
        if: steps.check-changes.outputs.changed == 'true'
        uses: peter-evans/create-pull-request@v6
        with:
          token: ${{ secrets.GITHUB_TOKEN }}
          commit-message: 'chore: update timezone data'
          title: '🔄 Automated Timezone Data Update'
          body: |
            ## Automated Timezone Data Update

            This PR contains the latest timezone data from [countries-states-cities-database](https://github.com/dr5hn/countries-states-cities-database).

            ### Statistics
            - Timezones: ${{ steps.data-stats.outputs.timezones }}
            - Countries with timezones: ${{ steps.data-stats.outputs.countries }}
            - Abbreviations: ${{ steps.data-stats.outputs.abbreviations }}

            ### Changes
            - Downloaded latest dataset from source repository
            - Regenerated timezone data structure
            - Updated timezone abbreviations
            - All tests passing ✅

            ### Verification
            - ✅ Data generation completed successfully
            - ✅ Build successful
            - ✅ All tests passing

            This is an automated pull request. Please review the changes before merging.
          branch: automated-data-update
          delete-branch: true
          labels: |
            automated
            data-update

      - name: No changes detected
        if: steps.check-changes.outputs.changed == 'false'
        run: |
          echo "✅ Data is already up to date. No PR created."