Python

The default Python distribution for Ubuntu is installed. You can start it with python3. To keep the system as flexible as possible for all users, no extra packages have been installed except pip and venv. To use other packages, follow the instructions below:

Virtual Environment

With a virtual environment, you create an isolated Python environment in which you can install other packages easily. You can create as many virtual environments as needed, typically one per project. This allows you to install different versions of packages that might be specifically needed for an application or might conflict with other packages.

To create a virtual environment, you use the venv package. It is advisable to set up a new directory per virtual environment, preferably the root directory for your project. Furthermore, it can be very helpful to create a special text file containing all packages needed for a project. It is customary to use the filename requirements.txt, in which you list each package on a separate line.

As a bonus, there is an easy way to automatically start the virtual environment when entering the directory for your virtual environment. This can be done with the direnv application. When set up in your .bashrc, this application reads the contents of the file .envrc whenever you enter the directory or its subdirectories.

Create the Virtual Environment

For this instruction, we want to have a project directory that will contain our Python code. We would like to use numpy and matplotlib in our Python code:

mkdir project
cd project
python3 -m venv env # The directory `env` is created for the virtual environment.
source env/bin/activate # Activate the virtual environment; check your prompt!

Install Packages

You now have the env directory and have activated the virtual environment. You should see this indicated in your command line prompt with the addition (env). Now, we will install the packages:

# The following instruction creates the file `requirements.txt` with the contents "numpy" and "matplotlib".
cat << EOF > requirements.txt
numpy
matplotlib
EOF
pip install -r requirements.txt --upgrade

The pip command will read the requirements.txt file and install (or upgrade) the packages in the virtual environment (in the env directory).

Automate Activation

With direnv, the virtual environment is activated once you enter the directory. For this to work, you need to add the direnv hook to your .bashrc once:

echo 'eval "$(direnv hook bash)"' >> .bashrc

Now, every time you log in to the server, this hook will be activated. For now, activate it manually with this command:

eval "$(direnv hook bash)"

Now, set up the directory to automatically activate the virtual environment:

cd project
cat << EOF > .envrc
source env/bin/activate
unset PS1
EOF
direnv allow

Check if the (env) prompt changes when leaving and entering the directory with the .envrc file.

Deactivate Manually

If you do not use direnv, you can always deactivate the virtual environment with:

deactivate