pre-committed

Collection of dev environment setups with basic pre-commit/linting configs. Contains advanced options such as:

OS recognition and compatibility (Mac, Linux, Windows)
Language specific .gitignore creation with API call to gitignore.io
Python environment creation for venv, conda, mamba
Git init

Install cookiecutter in your Python environment or with your package manager of choice, e.g., with pip install cookiecutter or brew install cookiecutter.

Usage

Run, choose your template, and follow the instructions with:

cookiecutter https://github.com/jxhnx/pre-committed

The templates are usually build so that options like git init or environment creation with, e.g., venv or conda are only available if they are found on your system.

Prerequisites

This repo and its templates make use of pre-commit. Check your installation with pre-commit -V. You can install pre-commit in you Python environment of choice, e.g. with pip install pre-commit or globally with your favorite package manager, e.g. with brew install pre-commit.

Development

To create a new template, you can use the default_template directory as starting point. Copy it to templates-single/ or templates-stack/, and adjust the files to your needs. Finally, register your template in the root cookiecutter.json file. You can find more information about cookiecutter in its docs. Some template file types or files may have to be added (ignored) in the .prettierignore or excluded in the .pre-commit-config.yaml.

Templates are divided into two categories: single and stack. Single templates are for a single language, e.g. Python. Stack templates are for a combination of languages and frameworks, e.g. Python and Django. Furthermore, more specialized, use-case specific templates are denoted with _usecase1, _usecase2, etc.

├── templates-single/
│   ├── python/
│   ├── python_usecase1/
│   ├── python_usecase2/
│   ├── javascript/
│   └── ...
└── templates-stack/
    ├── python-django/
    ├── python-django_usecase1/
    └── ...

Testing

Tests are written with pytest. Since some tests will create Python environments, it is a good idea to run them in a Docker container. To run the tests, cd into tests and run, e.g.,:

docker compose up test-default

The default test-set includes template creation with default values and a pre-commit linting check on the created repos. They use @pytest.mark.default and are also run in GitHub's CI.