Data Update #3
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Continuous Integration workflow for automated monthly data updates | |
| name: Data Update | |
| on: | |
| # Schedule to run on the 1st day of every month at midnight UTC | |
| schedule: | |
| - cron: "0 0 1 * *" | |
| # Allow manual trigger from the Actions tab | |
| workflow_dispatch: | |
| jobs: | |
| update: | |
| # Use the latest Ubuntu runner | |
| runs-on: ubuntu-latest | |
| # Define environment variables | |
| env: | |
| name: python | |
| RELEASE_DATE: "" # Placeholder for release date, set dynamically during the job | |
| PROJECT_DIR: ${{ github.workspace }} # Set project directory to the current GitHub workspace | |
| steps: | |
| # Step 1: Checkout the repository | |
| - name: Checkout repository | |
| uses: actions/checkout@v4 | |
| with: | |
| fetch-depth: 1 # Fetch only the latest commit for efficiency | |
| # Step 2: Set up Python environment | |
| - name: Set up Python | |
| uses: actions/setup-python@v5 | |
| with: | |
| python-version: "3.x" # Use the latest stable Python 3.x version | |
| # Step 3: Set the release date for the update | |
| - name: Set release date | |
| run: | | |
| echo "RELEASE_DATE=$(date --rfc-3339=date)" >> ${GITHUB_ENV} | |
| # Step 4: Install dependencies from the requirements file | |
| - name: Install dependencies | |
| run: | | |
| echo "Installing Python dependencies..." | |
| pip install -r requirements.txt # Install required dependencies | |
| # Step 5: Execute Jupyter notebooks via a custom script | |
| - name: Run Jupyter notebooks | |
| run: | | |
| echo "Running all Jupyter notebooks..." | |
| bash ./PIPELINES/all_runner.sh | |
| # Step 6: Create a pull request with the data updates | |
| - name: Create Pull Request | |
| id: cpr # Set an ID to reference outputs | |
| uses: peter-evans/create-pull-request@v6 | |
| with: | |
| token: ${{ secrets.GITHUB_TOKEN }} # Use the repo's GitHub token for authentication | |
| commit-message: "Data update as of ${{ env.RELEASE_DATE }}" | |
| committer: github-actions[bot] <github-actions[bot]@users.noreply.github.com> | |
| author: ${{ github.actor }} <${{ github.actor_id }}+${{ github.actor }}@users.noreply.github.com> | |
| signoff: false | |
| base: ${{ github.ref_name }} # Create the PR against the main branch | |
| branch: data-update # Create a new branch for the updates | |
| delete-branch: true # Delete the branch after the PR is merged | |
| title: "[PanditBot] Updating Repo Data ${{ env.RELEASE_DATE }}" | |
| body: | | |
| This PR contains the automated data updates. | |
| - Updated as of *${{ env.RELEASE_DATE }}*. | |
| labels: | | |
| report | |
| automated-pr | |
| assignees: ptprashanttripathi # Assign the PR to yourself | |
| reviewers: ptprashanttripathi # Set yourself as the reviewer | |
| draft: false # Mark the PR as ready for review | |
| # Step 7: Log outputs for visibility (PR number and URL) | |
| - name: Log PR Details | |
| if: ${{ steps.cpr.outputs.pull-request-number }} | |
| run: | | |
| echo "Pull Request Number: ${{ steps.cpr.outputs.pull-request-number }}" | |
| echo "Pull Request URL: ${{ steps.cpr.outputs.pull-request-url }}" |