Back to the overview

Using environment variables to configure Django

My preferred way to read values from the environment uses the ast module to evaluate Python literals. This means that values such as ["*"], None and True aren’t returned as strings but actually have the expected type already.

A basic implementation using ast.literal_eval() follows:

import ast
import os

def env(key):
    value = os.environ[key]
    try:
        return ast.literal_eval(value)
    except (SyntaxError, ValueError):
        return value

Some deployment environments may not make it easy to add environment variables. I like placing a .env file into the project root containing additional key-value combinations. Here’s a way to add those values to the environment:

def read_speckenv():
    with open(".env") as f:
        for line in f:
            line = line.strip()
            if not line or line.startswith("#") or "=" not in line:
                continue
            key, value = [v.strip("'\" \t") for v in line.split("=", 1)]
            os.environ.setdefault(key, value)

Those two functions along with a few additional bells and whistles are available as speckenv from PyPI. The packaged version also supports defaults, warning messages, coercion and using another mapping instead of os.environ. The implementation follows the basics outlined above.

Of course it may be useful to bundle the conversion of DSNs to the configuration format Django expects, however I mostly use individual libraries such as dj-database-url and dj-email-url for this.


© 2005 - 2020 Matthias Kestenholz