Dynaconf

Introduction

Dynaconf is a powerful and flexible configuration management tool for Python that simplifies how you handle settings and secrets across multiple environments (dev, staging, production, etc.).

Key Features

Inspired by the 12-factor application guide
Settings management with defaults, validation, parsing, and templating
Protection of sensitive information like passwords and tokens
Supports multiple file formats: toml, yaml, json, ini, py
Environment variable overrides with dotenv support
Layered configuration for multi environments: default, development, testing, production
Built-in support for Hashicorp Vault and Redis for secrets management
Extensions for Django and Flask web frameworks
CLI for operations like init, list, write, validate, and export
Layered merging of settings from prioritized sources

Installation

uv add dynaconf

Configuration

Use Dynaconf in your Python code:

from dynaconf import Dynaconf

settings = Dynaconf(
    environments=True,
    envvar_prefix="DYNACONF",
    load_dotenv=True,
    settings_files=["configs/settings.toml", "configs/.secrets.toml"],
)

environments=True enables environment sections like [dev], [prod]
envvar_prefix=”DYNACONF” prefixes environment variables for overrides
load_dotenv=True loads variables from a .env file
settings_files is a prioritized list of config files loaded by Dynaconf

Usage

Select environment by setting:
```
export ENV_FOR_DYNACONF=dev
```

Access settings in code:

print(settings.app_name)
print(settings.db_url)

Switch environments in code:

Create new instance for environment (recommended):

prod_settings = settings.from_env("prod")
print(prod_settings.db_url)

Temporary context switch:

with settings.using_env("prod"):
    print(settings.db_url)

Additional Resources

Dynaconf Official Docs: https://www.dynaconf.com/
Secrets Management: https://www.dynaconf.com/secrets/
Advanced Usage: https://www.dynaconf.com/advanced/
12-factor Config Guide: https://12factor.net/config

Next Step

After setting up Dynaconf, the next step is to explore Fire, a CLI generation tool that turns Python code into command-line interfaces easily.

Uninstall

uv remove dynaconf