python-gcplisted

# Stack Profile: Python + GCP This profile is automatically loaded when working in a Python project targeting GCP. ## Architecture Conventions - **FastAPI** or **Flask** for HTTP services - **SQLAlchemy** or **Prisma** for database access - **Pydantic** for data validation and serialization - **pytest** for testing - **Poetry** or **pip** for dependency management - **Docker** for containerization, **Cloud Run** for deployment ## Project Structure ``` src/ ├── api/ # Route handlers │ ├── __init__.py │ └── routes/ ├── core/ # Config, settings, dependencies ├── models/ # SQLAlchemy/Pydantic models ├── services/ # Business logic ├── repositories/ # Database access layer └── main.py # Application entrypoint tests/ ├── unit/ ├── integration/ └── conftest.py ``` ## GCP Services - **Cloud Run**: stateless HTTP services, auto-scaling to zero - **BigQuery**: analytics warehouse, columnar storage, SQL interface - **Cloud SQL**: managed PostgreSQL for transactional data - **Pub/Sub**: async messaging between services - **Cloud Storage**: file/object storage (GCS) - **Secret Manager**: secrets, not environment variables for sensitive values - **Cloud Build**: CI/CD with `cloudbuild.yaml` - **Cloud Scheduler**: cron jobs triggering Cloud Run or Pub/Sub ## Key Patterns - Use Pydantic `BaseSettings` for environment variable management - Structured logging with `google-cloud-logging` or `structlog` - Health check endpoint