Python Environment Setup

I am a python developer and usually work with multiple repositories at any time. Most of these projects have different dependencies so I create a virtual environment in every project. I prefer creating virtual env right next to code. Pycharm also detects it when repository is opened in Pycharm.

This setup worked most of the time but sometimes, I activate venv of one repository and then change it another repository without changing activating correct environment. I usually spotted this when code execution used to fail and never lost more than a minute on this but, this always made me think what if there is a shell that could run something when you enter a directory?

I decided to implement something simple in shell script temporarily. I ended up with creating small function, aliased with cd and added in my .bashrc.

It's a simple function and all I ever needed. It does,

1. Loads the .venv from that repository
2. Additionally, sources .cdenv, i usually set env variable but you can do anything in there.

I was okay with not deactivating/resetting environment when i exited the directory.

Add below function in .bashrc and you should be able to use it.

Note: If you create your virtual environment with different name then please change the name in code below.

function cd() {
  # Run a shell `cd` command or error
  command cd "$@" || return

  # If .venv exists in the directory
  if [[ -d .venv ]]; then
    # Deactivate any initialized virtual env, ignore error
    deactivate > /dev/null 2>&1

    # activate .venv
    source .venv/bin/activate
    echo "deactivate prev venv and activate current venv"
  fi

  # Load a .cdenv file
  if [[ -f .cdenv ]]; then
      echo "loading .cdenv"
      source ./.cdenv
  fi
}

Comments

[Martin Baun]

PyCharm as your IDE is the bomb. Out of the box, creating a new project will create and enable a virtual environment for you. From there you just 'pip install' using the PyCharm terminal like you would anyway.
Thanks for putting that in detail!

[Sneha]

In fact, it can also be combined with Servbay for environment configuration, enabling one-stop installation and version control for Python, which directly improves work efficiency. For more details, please check my main article. Hope it can provide you with a new option!