archive.yml

name: Archive

on:
  schedule:
    - cron: "12 7 * * *"
  workflow_dispatch:

env:
  CI: true
  PIPENV_VENV_IN_PROJECT: true
  SCRAPY_SETTINGS_MODULE: city_scrapers.settings.archive
  AUTOTHROTTLE_MAX_DELAY: 30.0
  AUTOTHROTTLE_START_DELAY: 1.5
  AUTOTHROTTLE_TARGET_CONCURRENCY: 3.0
  OPENVPN_USER: ${{ secrets.OPENVPN_USER }}
  OPENVPN_PASS: ${{ secrets.OPENVPN_PASS }}
  OPENVPN_CONFIG: ${{ secrets.OPENVPN_CONFIG }}

jobs:
  crawl:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v1

      - name: Connect to OpenVPN
        run: |
          sudo apt-get install -y openvpn
          echo "$OPENVPN_USER" | sudo tee -a /etc/openvpn/client/auth
          echo "$OPENVPN_PASS" | sudo tee -a /etc/openvpn/client/auth
          echo "$OPENVPN_CONFIG" | sudo tee -a /etc/openvpn/ovpn.conf
          sudo openvpn --config /etc/openvpn/ovpn.conf --daemon
          sleep 120

      - name: Set up Python 3.9
        uses: actions/setup-python@v1
        with:
          python-version: 3.9

      - name: Install Pipenv
        uses: dschep/install-pipenv-action@v1

      - name: Install dependencies
        run: pipenv sync
        env:
          PIPENV_DEFAULT_PYTHON_VERSION: 3.9

      - name: Run scrapers
        run: |
          export PYTHONPATH=$(pwd):$PYTHONPATH
          ./.deploy.sh