Environment variables with Python

When we need to use secret data on our apps (such as API keys or whatever), we can set environment variables and then:

• using the os library (for most Python apps and Jupyter notebooks)
• using the google.colab.userdata library (for Google Colab notebooks)

Python apps and Jupyter notebooks

Setting the "secret" variables

We can set the secret variable up in two ways:

Operating system wide

On a Mac or on a Linux, running this command on the Terminal from any folder will set the variable for use across the entire system:

export SECRET_NAME="secret_value_here"

On Windows, just go to Command Prompt:

C:\> set SECRET_NAME=secret_value

This way works if we know that we will use this variable across almost every app but we should try the next way for project-specific secret variables!

On an .env file

For project-specific secrets, create a file named .env and fill it with as many variables as needed, e.g.:

SECRET_NAME_1="secret_value_1_here"
SECRET_NAME_2="secret_value_2_here"

Let's also not forget to include .env in the .gitignore file to keep the secrets, secret!

Using them in a Python file

Now, we can use them in our Python program (but not without the os library):

# apiapp.py

import os

# if using .env file
from dotenv import load_dotenv
load_dotenv()

secret = os.environ.get("SECRET_NAME")

# another example
api_key = os.environ.get("MY_SECRET_API_KEY")

# rest of program - happy coding!

Google Colab notebooks

In a Google Colab notebook, things happen more easily:

• click on the "key" icon on the left-side navigation menu
• click on "+ Add new secret"
• turn the "Notebook access" switch to the "on" position
• enter the secret name under "Name"
• enter the secret value under "Value"

Then, in the code:

from google.colab import userdata
my_secret = userdata.get('SECRET_NAME')

Google Colab will show us something like:

from google.colab import userdata
userdata.get('SECRET_NAME')

...but should show us that we can actually use it in a variable ;)