Python Polylith Example

This is a repository with an example Python setup of the Polylith Architecture. Here you will find examples of code being shared between different kind of projects, and the developer tooling setup.

Developer experience

Mypy

Have a look at the mypy.ini configuration file, to make Mypy work really well with this type of architecture.

[mypy]
mypy_path = components, bases
namespace_packages = True
explicit_package_bases = True

The "loose" theme

This repository is setup with the loose theme, a Python exclusive addition to the architecture. The theme is about the folder structure of components:

components/<namespace>/<name>, and a separate tests top folder.

Currently, there is poor support in Mypy for the original Polylith component structure:

components/<name>/<src or test>/<namespace>/<name>/

With the original Polylith structure, you will have to explicitly add each component path to the mypy_path. There is a feature request in the mypy repo about adding regex support to the mypy_path property. Hopefully it will be implemented in the future. 🙏

.venv

It is recommended to create the virtual environment locally, for a great code editor experience. By default, Poetry will create a venv outside of the repo. You can override that behaviour by adding a configuration in a poetry.toml file:

[virtualenvs]
path = ".venv"
in-project = true

Tooling support

There's tooling support for using Polylith in Python. Have a look at this repository: Python tools for the Polylith Architecture

Tests

Run tests with:

poetry run pytest

Deployment

The projects contain examples for packaging and deploying APIs and handlers with docker, AWS Lambdas and GCP functions.

Kubernetes

The "My FastAPI project" includes an example setup for deploying into a Kubernetes cluser, by using kustomize to configure the deployments.

Have a look at the project-specific and the shared configurations.

The configuration is structured as:

• basic setup for all types of deployments
  • with overlays for different environments (development, staging, production)
• service-specific setup, with overlays
  • APIs
  • handlers
• project-specific setup, with overlays