Python and VSCode 101

This is a fast start with Python using Visual Studio Code in Ubuntu 20.04 LTS.

Development environment

The python3-pip and python3-venv are required. Install them with:

sudo apt install python3-pip python3-venv

Project’s development environment

In the project directory, create, activate and test a virtual python environment:

# create a virtual environment
python3 -m venv .venv

# activate the virtual environment
source .venv/bin/activate

# print the version and the path of the activated environment
python3 -c "import sys; print(sys.version); print(sys.path)"

Launch VSCode from that directory (code .) and prepare the python project:

open the Command Palette (Ctrl+Shift+P)
execute the Python: Select Interpreter command
select the previously created python environment

VSCode should add:

    "python.pythonPath": ".venv/bin/python"

to the .vscode/settings.json file.

Development and production modes

By default the Python extension looks for and loads an environment variable definitions file named .env in the current workspace folder and applies the definitions from that file. The file is identified by the default entry "python.envFile": "${workspaceFolder}/.env" in user settings. For debugging there is an envFile property that also defaults to "${workspaceFolder}/.env".

Some environment variables (credentials, databases, …) are different in development and production mode.

Create the default .vscode/launch.json file (Run > Open configurations will do it) and add the envFile entry like this:

{
    "version": "0.2.0",
    "configurations": [
        {
            "name": "Python: Current File",
            "type": "python",
            "request": "launch",
            "program": "${file}",
            "console": "integratedTerminal",
            "envFile": "${workspaceFolder}/dev.env"
        }
    ]
}

Edit .vscode/settings.json and add the entry for prod.env like this:

{
    "python.pythonPath": ".venv/bin/python",
    "python.envFile": "${workspaceFolder}/prod.env"
}

Add /dev.env in .gitignore (you may not want to publish it !) then create the dev.env and prod.env files.

Packages

Create a new integrated terminal (Ctrl+Shift+`) - this will also activate the virtual environment. In this new terminal install the packages you need.

Example: For CalinRadoni repo I have used:

PyGitHub, a Python library to access the GitHub API
Jinja, a full-featured template engine for Python

python3 -m pip install PyGithub jinja2

To upgrade the packages use:

python3 -m pip install --upgrade

Project’s requirements

Create a new integrated terminal (Ctrl+Shift+`) - this will also activate the virtual environment.

To create the list of the packages required by the project use:

pip3 freeze > requirements.txt

Remember that this file have to be recreated after installing, updating or removing packages.

After cloning the repository, the requirements can be installed with:

pip3 install -r requirements.txt

For more information read the Using requirements files document.

Git pre-commit hook

By design the hooks are not under version control. The method presented here creates a links a script from the repository to a git hook.

Each time a repo is cloned the link must be created for it to be used.

Warning: the pre-commit script can be changed and it will be run automatically before commit. This is a safety and security risk. Check the script before executing git commit or do not link it and do not launch it !

To use a pre-commit hook under version control, create an executable script in a directory in your repository (let’s say in utils directory) and link it to git’s pre-commit hook:

ln -s -f ../../utils/pre-commit .git/hooks/pre-commit