Python Package Generator¶
What’s included?¶
Generator of Python Project (see Quickstart), with CLI for Linux, MacOS, and Windows
Option to Generate Python Package designed as module, module+cli, or pytest-plugin!
Scaffold over 24 files, from Template, to have a ready-to-develop Project equipped with:
Fully-featured CI/CD Pipeline, running on Github Actions, defined in .github/
Continuous Delivery to PyPI (i.e. pypi.org, test.pypi.org) and Dockerhub
Continuous Integration, with Test Suite running pytest, located in the tests dir
Continuous Documentation, building with mkdocs or sphinx, and hosting on readthedocs, located in the docs dir
Static Type Checking, using mypy
Lint Check and Apply commands, using the fast Ruff linter, along with standard isort, and black
Build Command, using the build python package
What to expect?¶
You can to be up and running with a new Python Package, and run workflows on Github Actions, such as:
Link: https://github.com/boromir674/biskotaki/actions/runs/4157571651
CI Pipeline, running on Github Actions, defined in .github/
Job Matrix, spanning different platform’s and python version’s
Platforms: ubuntu-latest, macos-latest
Python Interpreters: 3.7, 3.8, 3.9, 3.10, 3.11, 3.12
Parallel Job execution, generated from the matrix, that runs the Test Suite
Artifact store of Source and Wheel Distributions, factoring Platform and Python Version
Auto Generated Sample Package Biskotaki¶
Check the Biskotaki Python Package Project, for a taste of the project structure and capabilities this Template can generate!
It it entirely generated using this Python Package Template:
Quickstart¶
To install the latest Generator in your environment, run:
pip install cookiecutter-python
The generate-python (executable) CLI should now be available in your environment.
Next, create a file, let’s call it gen-config.yml, with the following content:
default_context:
project_name: Demo Generated Project
project_type: module+cli
full_name: John Doe
email: john.doe@something.org
github_username: john-doe
project_short_description: 'Demo Generated Project Description'
initialize_git_repo: 'no'
interpreters: {"supported-interpreters": ["3.8", "3.9", "3.10", "3.11"]}
To generate a Python Package Project, run:
mkdir gen-demo-dir
cd gen-demo-dir
generate-python --config-file ../gen-config.yml --no-input
Now, you should have generated a new Project for a Python Package, based on the Template!
The Project should be located in the newly created demo-generated-project directory.
To leverage all out-of-the-box development operations (ie scripts), install tox:
python3 -m pip install --user 'tox==3.27.1'
To verify tox available in your environment, run: tox --version
Please, do a cd into the newly created directory, ie cd <my-great-python-package>.
To run the Test Suite, cd into the newly created Project folder, and run:
tox -e dev
All Tests should pass, and you should see a coverage report!
To run Type Checking against the Source Code, run:
tox -e type
All Type Checks should pass!
To setup a Git Repository, run:
git init
git add .
git checkout -b main
git commit -m "Initial commit"
To setup a Remote Repository, run for example:
git remote add origin <remote-repository-url>
git push -u origin main
To trigger the CI/CD Pipeline, run:
git push
Navigate to your github.com/username/your-repo/actions page, to see the CI Pipeline running!
Read the Documentation’s Use Cases section for more on how to leverage your generated Python Package features.
Next Steps¶
To prepare for an Open Source Project Development Lifecycle, you should visit the following websites:
PyPI, test.pypi.org, Dockerhub, and Read the Docs, for setting up Release and Documentation Pipelines
github.com/your-account to configure Actions, through the web UI
Codecov, Codacy, and Codeclimate, for setting up Automated Code Quality, with CI Pipelines
https://www.bestpractices.dev/ for registering your Project for OpenSSF Best Practices Badge
Happy Developing!