Python Environment Variables: Set a Variable in Bash

Avoiding Horror Scenarios

The web is full of horror stories of accidentally posted API key secrets that ended up costing the owner a lot of money. If you need some extra convincing, or just want to stay up late tonight, check out the following posts:

• A Git Horror Story: Repository Integrity With Signed Commits
• My $500 Cloud Security Screwup
• Dev put AWS keys on Github. Then BAD THINGS happened

The quick take-away is that you should never post your secrets to GitHub.

Bots are quick, and one compromised commit is one too many.

There are multiple ways to keep your sensitive information safe. In this article, you’ll learn how to do it using Python and environment variables in UNIX systems.

Using Environment Variables In Bash

Environment variables are dynamic-named values, which you can access from anywhere in your current environment. They can help you make running your scripts more user-friendly and secure, and are shared across all applications in your current environment.

In UNIX systems, the most famous one of them is $PATH, which specifies file paths where your system looks for executable files.

You can access the value of your environment variables anywhere in your project without ever spelling out the actual value of that variable. Instead, you can refer to it through the environment variable defined in Bash.

That way you can work with API secrets and passwords throughout your project, and commit all project-relevant code to GitHub while keeping your sensitive information safe and to yourself.

Inspect Environment Variables

Open up your CLI and type the Bash command printenv. This will give you a list of all the current environment variables present on your system:

This example output shows you a couple of environment variables that are currently defined on your local machine. You’ll probably see a different name and some additional lines in your own output.

You can check the value of each variable with echo $<NAME>. Because the environment variable $LOGNAME points to Martin in my case, I can confirm this using the echo command:

The output you receive when running the same command will be what your LOGNAME variable points to. You can confirm this value also by running printenv and looking for LOGNAME in your output.

With these two commands, you can inspect all the environment variables that are currently defined in your system. But what if you want to change, add, or remove one using Bash?

Set A Variable In Bash And Remove It

Using Bash in your CLI, you can add a new environment variable with the following command:

In this example, you’ll have to replace <NAME> with the new environment variable that you want to add. You also need to replace <VALUE> with the value you want to assign to that environment variable.

For example, to add a new variable with the name DAY and the value Sunday, you would spell it out as follows:

After executing this command, you can see that a new variable has been added to your environment. Take a look at it using printenv and echo as described above.

In order to remove an environment variable with Bash, you’ll have to call the following command:

Again, you’ll have to add the actual name of the variable you want to remove instead of the placeholder <NAME>:

This Bash command removes the DAY variable you set before.

Try adding and removing some environment variables using these Bash commands. Remember you can always check what’s happened using printenv or echo <NAME>.

However, when you are working on a Python web development project, you don’t want to set your environment variables across your whole system environment. For example, as soon as you’re working on more than one Django project, the SECRET_KEY variables you need for each project will clash with each other. That’s why you should compartmentalize your environment variables using virtual environments.

Using Environment Variables In Python Virtual Environments

When doing any project-specific development, you always want to avoid setting anything for your whole system. The same counts for secrets, which are usually project-specific.

For a project-specific secret, it’s a much better idea to set a variable in Bash inside of a virtual environment, which makes the variables turn into virtual environment variables.

Using environment variables inside of a Python virtual environment is easier than having to first export and then unset each variable every time that you want to work on a project.

Edit Your Activation Script

Start by creating a virtual environment for your Python project:

This command creates a virtual environment in your current project folder. You can learn more about working with virtual environments and Python in the Python Engineering course.

After you successfully created a virtual environment, open the activate script in your favorite text editor. You can find this file inside of the venv folder that got created by running the command shown above. The relative path of this script is venv/bin/activate.

This Bash script runs every time your venv gets activated, which makes it a good place to let your computer know which environment variables you would like to have, and which ones to get rid of once you exit the virtual environment.

You will now edit this script and set the variable in Bash that you want in the environment. First, you need to make sure that your virtual environment variables won’t stick around once you deactivated them, so you start by unsetting the environment variable that you haven’t even created yet.

Unset Virtual Environment Variables

To unset a virtual environment variable, you add an unset command to the deactivate command section in your Bash activate script. The code in this part of the script runs every time you deactivate your virtual environment.

For example, to unset the variable MY_SUPER_SECRET_SECRET, you need to add the following line of Bash code:

Adding this line in the deactivate section makes sure that your virtual environment variables won’t leak into your system environment. Instead, they’ll exist only within your virtual environment.

Once you wrote the code to unset your variable, it’s time to make sure you also set it, so it’ll exist in the first place.

Set Virtual Environment Variables

You can set a virtual environment variable in the same way as you practiced before when using the Bash CLI for it. However, instead of typing the export command directly in your terminal, you’ll add it as a new line of code at the end of the activate script:

Save the Bash script and close it. Now you can activate your virtual environment:

Once the virtual environment has been successfully activated, you can now run the printenv command to inspect the state of your environment variables in your current environment.

MY_SUPER_SECRET_SECRET should show up, as should the value you assigned to it.

After you confirmed that activating your virtual environment brings your virtual environment variable into existence, go ahead and deactivate it. Use printenv once again. Your secret should be gone.

As you can see, this setup can keep your project-specific secrets safe within their own comfy virtual environments.

Conclusion

Some secrets are meant to stay secret. You can set a variable in Bash and then use it in Python. Environment variables that you access from within your Python virtual environments allow you to work with project-specific secrets without running into the danger of accidentally committing them to public version control.

Make sure that you add your virtual environment folder to your .gitignore file, or you’ll end up pushing your secrets to GitHub after all!

In this blog post you learned how to:

• Set a variable in Bash, and remove it
• Set up project-specific environment variables inside of your Python virtual environments
• Access environment variables in Python as if they were Python environment variables

If you’re interested in learning about Python web development from the ground up and drilling best practices right from the start, check out CodingNomads’ courses on Python Engineering and Django Web Development.