Contents

Python environment

Python environment

The code you work on will run on a Python environment. By default, Python uses the system environment. It is recommended to create virtual environments for different projects to avoid library conflicts.

There are several ways to create/deal with Python environments:

  • Instantiate them manually using venv. In that case, VS Code won’t be able to find them automatically.

  • Manage Python interpreters and environments using Conda, in which case VS Code will find Conda environments automatically.

  • Use Poetry to manage a Python project. This option is fantastic for more mature projects that will later be released on PyPI. Poetry manages the project dependencies in a .venv/ subfolder; the dependencies are listed in the main pyproject.toml file.

The Python environment you select will be used for new terminal windows, running and debugging code.

If Python/Conda is not detected automatically

If required by your Python/Conda installation, update the following settings in the user settings.json, or through the UI:

    "python.defaultInterpreterPath": "/path/to/python",
    "python.condaPath": "/path/to/conda",

Selecting a Python environment in VS Code

CTRL+SHIFT+P Python: Select interpreter should show a list of autodetected environments. One can always select a venv that is not in the list.

This is the only VS Code setting that:

  • won’t be stored in the user settings.json file: indeed, the Python environment changes project by project

  • won’t be stored in the project .vscode/settings.json file: indeed, paths to Python environments change from computer to computer

Installing development dependencies

Below we list the tools that are required during development. Those tools are not required to use the Python packages you’ll develop.

Installation in a regular venv or in Conda

In a standard Python venv or one managed by Conda, use the commands:

pip install isort
pip install black
pip install mypy

(even though some/all these packages are available in Conda, the PyPI versions tend to be more up to date)

Installation using Poetry

If using Poetry, use the --dev argument to flag those dependencies as not required for simple use of the package you may publish later.

poetry add --dev black
poetry add --dev isort
poetry add --dev mypy

Installing Jupyter notebooks support

The Python VS Code extension installs support for Jupyter notebooks by default, see our explanation.

However, one needs to install the jupyterlab PyPI package to use that support.

pip install jupyterlab

or

poetry add --dev jupyterlab