1 line
5.3 KiB
JSON
1 line
5.3 KiB
JSON
{"repo": "alfonsocifuentes/riskmap", "files": {".github/workflows/ingest.yml": {"35743514b830325a3ad1cb9c8362487be07508b5": {"t": 1785747051.7963758, "n": 1610, "h": "1ce4823c33178e96", "c": "# ================================================================\n# RiskMap — Data Ingestion Pipeline (runs on GitHub Actions)\n# ================================================================\n# Ingests RSS feeds + NewsAPI → NLP analysis → stores in Neon Postgres\n# Schedule: every 2 hours\n# ================================================================\n\nname: \"📡 Data Ingestion Pipeline\"\n\non:\n schedule:\n - cron: '0 */2 * * *' # every 2 hours\n workflow_dispatch: # manual trigger\n\nenv:\n DATABASE_URL: ${{ secrets.DATABASE_URL }}\n NEWSAPI_KEY: ${{ secrets.NEWSAPI_KEY }}\n GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}\n DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}\n\njobs:\n ingest:\n runs-on: ubuntu-latest\n timeout-minutes: 25 # stay within free tier\n\n steps:\n - uses: actions/checkout@v4\n\n - uses: actions/setup-python@v5\n with:\n python-version: '3.11'\n cache: 'pip'\n\n - name: Install dependencies\n run: |\n pip install --upgrade pip\n pip install psycopg2-binary requests feedparser newspaper3k \\\n beautifulsoup4 lxml transformers torch sentencepiece \\\n groq deepseek-sdk geopy python-dotenv pydantic\n\n - name: Init schema\n run: python -m src.database.schema_init\n\n - name: Run ingestion\n run: python -m src.pipeline.ingest\n timeout-minutes: 15\n\n - name: Run NLP enrichment\n run: python -m src.pipeline.enrich\n timeout-minutes: 8\n\n - name: Run retention cleanup\n run: python -m src.database.retention\n"}, "db66499a43f1d5b86c2a53e63186e3a8b71fe91e": {"t": 1785770121.190131, "n": 1610, "h": "1ce4823c33178e96", "c": "# ================================================================\n# RiskMap — Data Ingestion Pipeline (runs on GitHub Actions)\n# ================================================================\n# Ingests RSS feeds + NewsAPI → NLP analysis → stores in Neon Postgres\n# Schedule: every 2 hours\n# ================================================================\n\nname: \"📡 Data Ingestion Pipeline\"\n\non:\n schedule:\n - cron: '0 */2 * * *' # every 2 hours\n workflow_dispatch: # manual trigger\n\nenv:\n DATABASE_URL: ${{ secrets.DATABASE_URL }}\n NEWSAPI_KEY: ${{ secrets.NEWSAPI_KEY }}\n GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}\n DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}\n\njobs:\n ingest:\n runs-on: ubuntu-latest\n timeout-minutes: 25 # stay within free tier\n\n steps:\n - uses: actions/checkout@v4\n\n - uses: actions/setup-python@v5\n with:\n python-version: '3.11'\n cache: 'pip'\n\n - name: Install dependencies\n run: |\n pip install --upgrade pip\n pip install psycopg2-binary requests feedparser newspaper3k \\\n beautifulsoup4 lxml transformers torch sentencepiece \\\n groq deepseek-sdk geopy python-dotenv pydantic\n\n - name: Init schema\n run: python -m src.database.schema_init\n\n - name: Run ingestion\n run: python -m src.pipeline.ingest\n timeout-minutes: 15\n\n - name: Run NLP enrichment\n run: python -m src.pipeline.enrich\n timeout-minutes: 8\n\n - name: Run retention cleanup\n run: python -m src.database.retention\n"}, "6a3d985937dfbbcd5162e8ac4dda8808747abd66": {"t": 1785770122.53856, "n": 1587, "h": "27aa00aa06f2ffd3", "c": "# ================================================================\n# RiskMap — Data Ingestion Pipeline (runs on GitHub Actions)\n# ================================================================\n# Ingests RSS feeds + NewsAPI → NLP analysis → stores in Neon Postgres\n# Schedule: every 2 hours\n# ================================================================\n\nname: \"📡 Data Ingestion Pipeline\"\n\non:\n schedule:\n - cron: '0 */2 * * *' # every 2 hours\n workflow_dispatch: # manual trigger\n\nenv:\n DATABASE_URL: ${{ secrets.DATABASE_URL }}\n NEWSAPI_KEY: ${{ secrets.NEWSAPI_KEY }}\n GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}\n DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}\n\njobs:\n ingest:\n runs-on: ubuntu-latest\n timeout-minutes: 25 # stay within free tier\n\n steps:\n - uses: actions/checkout@v4\n\n - uses: actions/setup-python@v5\n with:\n python-version: '3.11'\n cache: 'pip'\n\n - name: Install dependencies\n run: |\n pip install --upgrade pip\n pip install psycopg2-binary requests feedparser newspaper3k \\\n beautifulsoup4 lxml transformers torch sentencepiece \\\n groq deepseek-sdk geopy\n\n - name: Init schema\n run: python -m src.database.schema_init\n\n - name: Run ingestion\n run: python -m src.pipeline.ingest\n timeout-minutes: 15\n\n - name: Run NLP enrichment\n run: python -m src.pipeline.enrich\n timeout-minutes: 8\n\n - name: Run retention cleanup\n run: python -m src.database.retention\n"}}}} |