How to Run Different Versions of Python From Your Terminal.

Summary

The website content provides a guide on setting up version-specific virtual environments to run different versions of Python for different projects, with a focus on using symlinks for easy access to specific Python versions.

Abstract

The article discusses a solution for developers who need to use different Python versions for various projects. It highlights the case of requiring an older version of Python, such as 3.7, to install and run a library like Kivy, which may not support newer Python versions. The guide walks through the steps of installing the needed Python version, creating a symbolic link (symlink) for easy access, and setting up a virtual environment with the specified Python version using the virtualenv command with the -p flag. It also provides instructions for activating the virtual environment on Unix/Linux and Windows systems, ensuring that the correct Python version is used within the environment. The article concludes with helpful tips for managing virtual environments and promotes an AI service as a cost-effective alternative to ChatGPT Plus.

Opinions

The author expresses the practicality of using version-specific virtual environments as an "elegant solution" to managing different Python versions.
Kivy is described as a "super slick" library for front-end app development that supports Material Design, implying a positive opinion about its capabilities and design.
The author suggests that finding the right answers on forums can be like a "real treasure hunt," possibly indicating a preference for more direct and clear documentation or guides.
The recommendation of ZAI.chat as an AI service implies that the author believes it to be a valuable and cost-effective tool for developers, comparable to ChatGPT Plus.

In order to install the package I required, I needed to switch to an older version of Python, but only for that project. Here is the simple solution: Version specific virtual environments.

As of writing this, I am using the latest version of Python on my machine: 3.8.5. It is usually fine to use the latest version. Except every now and then, something juicy requires an earlier version of Python for it to work, or in some cases, it won’t even install without that older version.

In this case, I’m talking about Kivy. It’s a super slick front end app development library for Python. It lets you develop mobile apps for your device completely in Python, and it’s really solid from what I can see (it even supports Material Design). But for you, it could be anything that doesn’t have support for your version of Python. Here’s a really simple and elegant way to get your specific version up and running.

Install it on your system. Ok, you can’t use a specific version if it doesn’t exist on your system. So install the specific Python version you need. In this case, it was Python 3.7 and I already had it on my machine.

Next, find or create a symlink to it. When you type python in terminal, press tab before you press enter to see what you have available. You’ll see the symlinks (like shortcuts) to the different versions of Python that you have already.

If you see your python version in the list, skip to step 6. If not, you need to find out where it’s installed.

I had a version of Python 3.7 installed in /usr/bin/python3. (I figured this out from VSCode, when I went to change my python interpreter by selecting this little button here, it showed me the path).

List of installed Python interpreters and their locations in VS Code. You can get to this list by clicking the currently selected interpreter at the bottom left of the screen (where it says “Python 3.7.3 64-bit”, or by typing in “Python: Select Interpreter” from the command pallete.

5. So I created a symlink to it using the command:

ln -s /usr/bin/python3 /usr/local/bin/python37

Why did I need a symlink? For starters, a symlink lets me run Python 3.7 any time I want by typing in python37, but secondly, I wanted to be able to easily use it in my virtual environment.

If you don’t know about virtual environments, check my article on what they are and how to easily create one.

6. Now that I had my symlink, I simply navigated to my project folder that needed python python 3.7, and created my virtual environment, except this time, with the -p tag which allowed me to specify my python version for anything in this venv instance. Type:

virtualenv venv -p python37

7. Next, you want to activate your new virtual environment so, make sure you’re in the parent folder and type:

# For Unix/Linux based, type:
source venv/bin/activate

# For Windows, type:
. venv/Scripts/activate

8. Now type python --version to check if it worked. While you’re inside your virtual environment, you can type python, rather than your new symlink python37, because it is your primary version for all projects in this environment.

9. From this point, I was able to install Kivy without getting any weird errors, and everything worked great.

How to Run Different Versions of Python From Your Terminal.

Version specific virtual environments. The elegant solution to needing different Python versions for different projects.

In order to install the package I required, I needed to switch to an older version of Python, but only for that project. Here is the simple solution: Version specific virtual environments.

That’s it!

Helpful tips