Easy injection of environment variables from .env files.
If you spend a lot of time switching between multiple Python projects and using several virtual environment, you’ll be familiar with the pain of injecting the right environment variables into your scripts. You can use autoenv, but it doesn’t automatically load the virtual environment for you, so you need to litter the .env file with bash commands. Or you can use a postactivate hook in virtualenvwrapper, but that would mean not having a standardized, default location for all the needed environment variables. And what if you use an IDE like PyCharm, which doesn’t load variables from .env files?
Envious solves the problem by loading configurations from an .env file in the working directory directly from within your Python code, injecting environment variables if they’re not already defined, while keeping those that are already set.
To install envious, simply type:
$ pip install envious
To have your project import environment variables from an .env file, first create the file in the root folder of your project:
ENVIRONMENT=development MONGODB_URL=http://localhost:27017/mydb REDIS_URL=http://localhost:6379/0
Then add the following to the Python app entry point:
from envious import load_env load_env()
If you want to manage multiple configuration files for the same project, you’re covered. Create multiple configuration files, and then specify the one you want to use by providing an environment variable named ENV_FILE:
$ echo MONGODB_URL=http://localhost:27017/myseconddb > .env2 $ ENV_FILE=.env2 python my_script.py
The script will see the alternative value for the environment value MONGODB_URL.