scraper.yml

name: Scraper

on:
  schedule:
    - cron:  '0 5 * * MON'
  workflow_dispatch:

jobs:
  scrape:
    runs-on: ubuntu-latest
    steps:
    - name: Generate a token
      id: generate-token
      uses: actions/create-github-app-token@v1
      with:
        app-id: ${{ secrets.SCRAPER_APP_ID }}
        private-key: ${{ secrets.SCRAPER_APP_PRIVATE_KEY }}

    - uses: actions/checkout@v4
      with:
        token: ${{ steps.generate-token.outputs.token }}

    - uses: ruby/setup-ruby@v1
      with:
        bundler-cache: true

    - name: Run scraper
      env:
        GH_TOKEN: ${{ steps.generate-token.outputs.token }}
      run: |
        bundle exec rake scraper
        sed -i -e 's/ $//' db/data/*.yml
        git add db/data
        if ! git diff --cached --quiet;
        then
            git config --global user.email "micarrera@cedarcode.com"
            git config --global user.name "MiCarrera"
            git commit -m 'chore: update scraped subjects data'
            git push -f origin HEAD:scraper
            if [ $(gh pr list --head scraper --json title --jq 'length') == '0' ];
            then
                gh pr create \
                    --title "Update scraped subjects data" \
                    --body "This PR runs `rails scraper` to update scraped subjects' data" \
                    --head scraper
            else
                echo "A PR already exists and was correctly updated."
            fi
        else
            echo "There were no changes. Nothing to do."
        fi