Manage Environment Variables in Python Projects

I find I have to tell developers (sometimes even developers with a few years’ experience) again and again how to properly manage environment variables - so I write this simple post so that I can just send a link.

Don’ts

Don’ts: DO NOT ever write the keys in your code - you will forget and push the keys to your repo.

Dos

Do one of the followings (I use OPENAI_API_KEY as an example and my preferred way is method 2):

1. use export command

• Open terminal and run export OPENAI_API_KEY=sk-xxxxxxx

• Test that key is correctly set as an environment variable

    import os
  
    if os.getenv("OPENAI_API_KEY") is not None:
        print("OPENAI_API_KEY is ready")
    else:
        print("OPENAI_API_KEY environment variable not found")
  

1. use .env file

• create a .env file in the root of the project folder to store the environment variables - this file is gitignored by default (or add this file to .gitignore)

  

• create requirements.txt and include packages to install by putting the following lines in the file (two packages as examples):

    openai
    python-dotenv
  

• create a virtual environment and install the packages specified in requirements.txt:

    python -m venv venv
    source venv/bin/activate
    pip install -r requirements.txt
  

• load keys from .env file:

    import openai
    import os
    from dotenv import load_dotenv
  
    load_dotenv()
    openai.api_key = os.getenv('OPENAI_API_KEY')
  

Method 2 above is my preferred way especially when you have many environment variables to manage.

PS. The featured image for this post is generated using HiddenArt.ai.