Settings
Any Turbulette project created using the turb CLI comes with a default settings module with minimal settings to get a working hello world.
Settings management is another Django-inspired concept, but Turbulette extends it a bit by allowing applications to have their own settings module.
This avoid rewriting a settings registry when writing an application, and is an easy way to define default values.
The project settings module is always loaded last on startup, so default values can be set by applications (in their own settings module) and overridden in project settings.
Load settings from .env file
Following twelve-factor methodology , Turbulette strongly encourages strict separation between code and settings because config is likely to vary between deployments, code does not.
To help you, Starlette provides a Config object that reads settings from environment variables and/or .env files:
# settings.py
from starlette.config import Config
# Config will be read from environment variables and/or ".env" files.
config = Config(".env")
DEBUG = config('DEBUG', cast=bool, default=False)
See Starlette's Config docs for more details.
Try different paths
Turbulette has a small helper to try different paths when loading .env files:
# settings.py
from turbulette.conf import get_config_from_paths
config = get_config_from_paths(["project/.env", "../project/.env"])
get_config_from_paths will try all paths in order, stop on the first existing one and return a Config instance with it.
In the above example, you can start your server in the root folder (containing the the Turbulette project folder), or inside the project folder (in this case, the second path will be be picked).