Python virtual environments

Python lets you reuse code written by other developers. These reusable collections of code are called libraries. For example, installing FastAPI gives you ready-made functionality for building web APIs.

Installing every library into one shared Python environment can cause conflicts when different projects need different versions. Some Python distributions, including Homebrew’s, also protect their global environment from direct package installation.

Instead, create a virtual environment for each project. Starting a Python project by creating its virtual environment is standard practice.

Creating a virtual environment

Python includes a module named venv for this purpose. Create a project folder, enter it, and create an environment named .venv.

mkdir playground
cd playground
python3 -m venv .venv

mkdir playground
cd playground
py -m venv .venv

The final command creates a .venv folder containing the project’s Python environment and its own library installation directory.

Activating a virtual environment

Activate the environment to make python and pip in your terminal use the project’s environment rather than global Python.

source .venv/bin/activate

.venv\Scripts\Activate.ps1

Your prompt will normally gain a (.venv) prefix:

(.venv) playground $

Python now runs in the project’s isolated environment. Installing libraries here will not change the libraries in other projects’ environments.

Deactivating a virtual environment

When you are finished, return to the previous environment with the same command on either operating system:

deactivate

Exercises

Decide whether each statement is True or False.

1. Creating a virtual environment is a common first step in a new Python project.
2. After activation, the python command still uses the global Python environment.
3. deactivate ends the active virtual environment session.