FastAPI: Managing Environment-Specific Settings

WHY IT EXISTS: Applications need to run in different environments (local, staging, production) which require different configurations, like database URLs or API keys. Hardcoding these values is insecure and inflexible. A systematic way to manage these settings is needed to prevent errors and security leaks.

THE MENTAL MODEL: Think of your application's configuration as a typed contract. You declare all the external settings your app needs in a single Pydantic BaseSettings class. This class acts as a single source of truth, defining the setting's name, type, and default value. FastAPI then automatically populates this object from environment variables or a .env file.

HOW IT WORKS: You create a class that inherits from pydantic_settings.BaseSettings. Inside, you define attributes with type hints, like database_url: str. Pydantic Settings reads environment variables that match your attribute names (case-insensitively). For example, an environment variable DATABASE_URL will populate the database_url attribute. It can also automatically load these from a .env file if one is present. This gives you type-validated, centrally-managed settings that raise an error on startup if a required value is missing.

WHEN TO USE IT: Use this for any value that changes between deployments. This includes database connection strings, external API keys, secret keys for signing tokens, log levels (e.g., 'DEBUG' vs. 'INFO'), and feature flags. It's standard practice for any non-trivial FastAPI application.

WHEN NOT TO USE IT: Avoid using it for true application constants that never change, like a mathematical constant or a fixed algorithm parameter. These can be defined directly in your code as uppercase constants. Overusing it for static values can make the code harder to follow by hiding simple constants behind an environment loading mechanism.

ONE CANONICAL EXAMPLE: First, create a config.py file and define a settings class: from pydantic_settings import BaseSettings followed by class Settings(BaseSettings): app_name: str = "My API"; admin_email: str; database_url: str; class Config: env_file = ".env". Then, instantiate it: settings = Settings(). Next, create a .env file in your project root with the content: ADMIN_EMAIL=dev@example.com and DATABASE_URL=postgresql://user:pass@localhost/db. Now, in your main app, you can from config import settings and access validated values like settings.admin_email. Pydantic raises an error if required settings are missing.