Skip to content

Process URLs_V2

Process URLs_V2 #2037

name: Process URLs_V2
on:
workflow_dispatch:
inputs:
startIdx:
description: 'Start index'
required: true
default: '0'
endIdx:
description: 'End index'
required: true
default: '100'
jobs:
process_urls:
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
- name: Set up Python 3.8
uses: actions/setup-python@v2
with:
python-version: 3.8
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install -r requirements.txt
- name: Run script
env:
START_IDX_V2: ${{ github.event.inputs.startIdx }}
END_IDX_V2: ${{ github.event.inputs.endIdx }}
run: python scraping_data_V2.py
- name: Commit and push changes
run: |
git config --global user.name "${{ secrets.USERNAME_GITHUB }}"
git config --global user.email "${{ secrets.EMAIL_GITHUB }}"
git pull
git add .
git commit -m "Commit from GitHub Actions" || echo "No changes to commit"
git push
- name: Calculate new indices
id: calculate-indices
run: |
echo "::set-output name=newStartIdx::${{ github.event.inputs.endIdx }}"
echo "::set-output name=newEndIdx::$((${{ github.event.inputs.endIdx }} + 100))"
- name: Dispatch next workflow
if: ${{ github.event.inputs.endIdx < 203905 }} # or whatever your max value is
uses: benc-uk/workflow-dispatch@v1
with:
workflow: Process URLs_V2
token: ${{ secrets.YOUR_PERSONAL_ACCESS_TOKEN }}
inputs: '{"startIdx": "${{ steps.calculate-indices.outputs.newStartIdx }}", "endIdx": "${{ steps.calculate-indices.outputs.newEndIdx }}"}'