Skip to content

Using Frameworks

Stela loads at module import time when you use from stela import env. Because of that, it works seamlessly with many Python packages and frameworks. Below are recommendations and best practices:

Please check the folder: /examples

It contains working implementations for many frameworks.

Django

The best practice for Django projects is to consume the Stela env object on django settings.py module. This is because many django packages look at this file to consume and define his own settings.

# Django settings.py
from stela import env

DEBUG = env.DEBUG

Pydantic 2.x

Pydantic uses the BaseSettings logic to handle environment variables. If you want to use Pydantic Settings with Stela dotenv file combinations (.env, .env.local, etc..) you can use:

# .env.local
FOO__BAR=123
Use class StelaConfigSettingsSource to replace the original env_settings in your Settings class:

from typing import Tuple, Type
from pydantic_settings import (
    BaseSettings,
    PydanticBaseSettingsSource,
    SettingsConfigDict,
)
from stela.helpers.pydantic import StelaConfigSettingsSource


class FooSettings(BaseSettings):
    bar: str


class Settings(BaseSettings):

    model_config = SettingsConfigDict(extra="ignore", env_nested_delimiter="__", log_stela_settings=True)

    foo: FooSettings

    @classmethod
    def settings_customise_sources(
        cls,
        settings_cls: Type[BaseSettings],
        init_settings: PydanticBaseSettingsSource,
        env_settings: PydanticBaseSettingsSource,
        dotenv_settings: PydanticBaseSettingsSource,
        file_secret_settings: PydanticBaseSettingsSource,
    ) -> Tuple[PydanticBaseSettingsSource, ...]:
        return (
            init_settings,
            StelaConfigSettingsSource(settings_cls),
            file_secret_settings,
        )


print(Settings().foo.bar)
#> 123

FastAPI

You can use FastAPI with Stela directly ou via Pydantic Settings (if you use the settings above).

AWS Chalice

We currently don't support Chalice autoreload in local server. To work around this please use the command chalice local --no-autoreload or import env inside a function.

Jupyter Notebooks

The recommended practice is to import Stela data at the first cell, using the %set_env variables to define Stela behavior and then call read_env function to retrieve the Environment Variables:

# In a Jupyter cell (IPython magic):
%set_env STELA_ENV=remote
%set_env STELA_SHOW_LOGS=False

# In the next cell, reload and use Stela
from stela.utils import read_env

env = read_env()
print(f"Current Environment: {env.current_environment}")
print(f"My Secret: {env.MY_SECRET}")

For the next step, let's review all Stela options