Big Scrape (F25) #6
This file contains hidden or 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
| name: Big Scrape (F25) | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| dry_run: | |
| description: 'Dry run (1 account only)' | |
| required: false | |
| default: 'true' | |
| type: boolean | |
| limit: | |
| description: 'Max posts per user to scrape' | |
| required: false | |
| default: '100' | |
| type: string | |
| jobs: | |
| big_scrape: | |
| runs-on: ubuntu-latest | |
| permissions: | |
| contents: write | |
| env: | |
| # --- Django & App Config --- | |
| PRODUCTION: '1' | |
| DJANGO_SETTINGS_MODULE: 'config.settings.development' | |
| SECRET_KEY: ${{ secrets.SECRET_KEY }} | |
| CLERK_SECRET_KEY: ${{ secrets.CLERK_SECRET_KEY }} | |
| # --- Database Config --- | |
| DATABASE_URL: ${{ secrets.SUPABASE_DB_URL }} | |
| SUPABASE_DB_URL: ${{ secrets.SUPABASE_DB_URL }} | |
| POSTGRES_DB: ${{ secrets.POSTGRES_DB }} | |
| POSTGRES_USER: ${{ secrets.POSTGRES_USER }} | |
| POSTGRES_PASSWORD: ${{ secrets.POSTGRES_PASSWORD }} | |
| POSTGRES_HOST: ${{ secrets.POSTGRES_HOST }} | |
| POSTGRES_PORT: ${{ secrets.POSTGRES_PORT }} | |
| # --- Service Keys --- | |
| OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} | |
| AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }} | |
| AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }} | |
| AWS_S3_BUCKET_NAME: ${{ secrets.AWS_S3_BUCKET_NAME }} | |
| AWS_DEFAULT_REGION: ${{ secrets.AWS_DEFAULT_REGION }} | |
| RESEND_API_KEY: ${{ secrets.RESEND_API_KEY }} | |
| RESEND_FROM_EMAIL: ${{ secrets.RESEND_FROM_EMAIL }} | |
| EMAIL_ENCRYPTION_KEY: ${{ secrets.EMAIL_ENCRYPTION_KEY }} | |
| EMAIL_HASH_KEY: ${{ secrets.EMAIL_HASH_KEY }} | |
| # --- Apify Token --- | |
| APIFY_API_TOKEN: ${{ secrets.APIFY_API_TOKEN }} | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Set up Python | |
| uses: actions/setup-python@v5 | |
| with: | |
| python-version: '3.11' | |
| - name: Create logs directory | |
| working-directory: backend/scraping | |
| run: mkdir -p logs | |
| - name: Cache pip | |
| uses: actions/cache@v4 | |
| with: | |
| path: ~/.cache/pip | |
| key: ${{ runner.os }}-pip-${{ hashFiles('backend/requirements.txt') }} | |
| restore-keys: | | |
| ${{ runner.os }}-pip- | |
| - name: Install dependencies | |
| working-directory: backend | |
| run: | | |
| pip install --prefer-binary -r requirements.txt | |
| - name: Run scraper | |
| working-directory: backend/scraping | |
| run: | | |
| python -u main_big_scrape.py \ | |
| ${{ inputs.dry_run && '--dry-run' || '' }} \ | |
| --limit ${{ inputs.limit }} \ | |
| 2>&1 | tee logs/scraping.log | |
| continue-on-error: false | |
| - name: Upload logs as artifacts | |
| if: always() | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| name: logs-big-scrape-${{ github.run_number }} | |
| path: | | |
| backend/scraping/events_scraped.csv | |
| backend/scraping/scraping.log | |
| backend/scraping/apify_raw_results.json | |
| if-no-files-found: 'ignore' |