quatalog-scraper/.github/workflows/transfer.yml

name: Scrape transfer and update file
run-name: Scrape transfer and update file
on:
  workflow_dispatch:
  schedule:
    - cron: '0 10 */2 * *'

jobs:
  setup:
    name: Get number of pages and set up scrape page jobs
    runs-on: ubuntu-latest
    steps:
    - name: Create matrix parameters
      id: matrix-params
      run: |
        UA='User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:122.0) Gecko/20100101 Firefox/122.0'
        URL='https://tes.collegesource.com/publicview/TES_publicview01.aspx?rid=f080a477-bff8-46df-a5b2-25e9affdd4ed&aid=27b576bb-cd07-4e57-84d0-37475fde70ce'
        curl -s -S -H "$UA" "$URL" | tr -d '\0' > home.html
        echo "======================================================================"
        if [ -z "$(cat "home.html" | grep -A2 "Oops! Page currently not available (")" ]; then
          echo "Transfer guide unavailable"
          exit -1
        fi
        echo "======================================================================"
        cat "home.html" | grep -A2 "<title>"
        echo "======================================================================"
        NUM_PAGES="$(cat "home.html" | grep -e 'lblInstWithEQPaginationInfo' | grep -Poie '(?<=of )[0-9]*')"
        echo "Found $NUM_PAGES pages"
        MATRIX_PARAMS="$(seq -s "," 1 "$NUM_PAGES")"
        MATRIX_PARAMS="\"page\": $(sed -e 's/,/}, {"page": /g' <<< "$MATRIX_PARAMS")"
        echo "matrix-params={\"include\": [{"$MATRIX_PARAMS"}]}" | tee $GITHUB_OUTPUT
    outputs:
      matrix-params: ${{ steps.matrix-params.outputs.matrix-params }}

  scrape-page:
    name: Scrape page
    needs: setup
    uses: ./.github/workflows/scrape_page.yml
    strategy:
      matrix: ${{ fromJson(needs.setup.outputs.matrix-params) }}
      fail-fast: true
    with:
      page: ${{ matrix.page }}

  commit-data:
    name: Combine/convert and commit data
    runs-on: ubuntu-latest
    needs: scrape-page
    steps:
      - name: Set up python
        uses: actions/setup-python@v5
        with:
          python-version: '3.11'

      - name: Checkout data repo
        uses: actions/checkout@v4
        with:
          repository: quatalog/data
          path: data
          token: ${{ secrets.PUSH_TOKEN }}

      - name: Checkout scraping repo
        uses: actions/checkout@v4
        with:
          path: scrapers

      - name: Download partial JSONs
        uses: actions/download-artifact@v4
        with:
          pattern: transfer-page-*
          merge-multiple: true
          path: new-data

      - name: Combine JSONs
        run: |
          cat new-data/* | jq -s 'add | sort_by(.institution)' > data/transfer.json

      - name: Commit data
        working-directory: data
        run: |
          git config user.name "Quatalog Updater"
          git config user.email "github_actions@quatalog.com"
          git add transfer.json
          git commit -m "$(date)" || exit 0
          git push

      - name: Trigger CSV generator
        run: |
          curl -H "Accept: application/vnd.github.everest-preview+json" \
               -H "Authorization: token ${{ secrets.PUSH_TOKEN }}" \
               -X POST --data '{"event_type": "generate-csv"}' \
               'https://api.github.com/repos/quatalog/quatalog/dispatches'