Using Pyprefab¶

CLI¶

pyprefab is a command line interface (CLI) that requires only a few pieces of information to create the boilerplate for a Python package.

To see the available options:

pyprefab --help

Example CLI use¶

To create a Python package named holodeck in a directory named trek/code/holodeck:

$ pyprefab --name holodeck --author rbarclay \
--description "personal holodeck programs" \
--dir trek/code/holodeck --docs

╭──────────── Package Created Successfully ─────────────╮
│ ✨ Created new package holodeck in trek/code/holodeck │
│ Author: rbarclay                                      │
│ Description: personal holodeck programs               │
│ Documentation: trek/code/holodeck/docs                │
╰───────────────────────────────────────────────────────╯

The scaffolding for the new holodeck package is now in the directory you specified (in this case, trek/code/holodeck).

trek/code/holodeck
  ├── .github
  │ ├── dependabot.yaml
  │ └── workflows
  │     ├── ci.yaml
  │     └── publish-pypi.yaml
  ├── .gitignore
  ├── .pre-commit-config.yaml
  ├── .python-version
  ├── CHANGELOG.md
  ├── CONTRIBUTING.md
  ├── docs
  │ └── source
  │     ├── _static
  │     │ └── custom.css
  │     ├── CHANGELOG.md
  │     ├── conf.py
  │     ├── CONTRIBUTING.md
  │     ├── index.rst
  │     ├── readme.md
  │     └── usage.md
  ├── LICENSE
  ├── pyproject.toml
  ├── README.md
  ├── src
  │ └── holodeck
  │     ├── __init__.py
  │     ├── __main__.py
  │     ├── app.py
  │     └── logging.py
  └── test
      └── test_app.py

Interactive mode¶

If you don’t explicitly specify options, pyprefab will prompt for them:

$ pyprefab
Package name 🐍: holodeck
Package author 👤 [None]: rbarclay
Package description 📝 [None]: personal holodeck programs
Package directory 🎬 [/Users/rbarclay/code/holodeck]:
Include Sphinx docs? 📄 [y/N]: y

╭───────────────── Package Created Successfully ──────────────╮
│ Created new package holodeck in /Users/rbarclay/holode      │
│ Author: rbarclay                                            │
│ Description: personal holodeck programs                     │
│ Documentation: /Users/rbarclay/holodeck/docs                │
╰─────────────────────────────────────────────────────────────╯

Using the new package¶

Now you’re ready to use the pyprefab-generated scaffolding to build your own Python package.

Creating a dev environment for the new package¶

Follow the steps below to create a development environment for Python packages generated by pyprefab.

These directions use uv, but you can use your preferred tooling.

cd to the directory of the new Python package
Create a virtual environment and install the project dependencies:

$ uv sync

Test the project setupt:

$ uv run <your_package_name>
You should see log output stating that the project has been set up correctly.

For example:
2025-01-13 02:29:08 [info] project_test successfully created.
You can also run the tests:
$ uv run pytest

Previewing documentation¶

If your project includes the optional Sphinx documentation, make sure you can build and preview the docs before updating them:

$ uv run --group docs sphinx-autobuild docs/source docs/_build/html

The output of the above command provides a URL for viewing the documentation via a local server (usually http://127.0.0.1:8000).

The HTML pages are in docs/_build/html.
[sphinx-autobuild] Serving on http://127.0.0.1:8000
[sphinx-autobuild] Waiting to detect changes...

Adding the project to git¶

To create a new git repository for the project (this is optional):

$ git init
$ git add .
$ git commit -am "Initial commit"

Tip

If you use pre-commit, pyprefab’s boilerplate includes a baseline pre-commit-config.yaml configuration. To use it, make sure the project has been added to git (see above) and install the pre-commit hooks: pre-commit install

Hosting your project on GitHub¶

Code generated by pyprefab includes some extra boilerplate files for projects hosted on GitHub. There are located in the .github directory, which you can delete if you’re not using GitHub.

GitHub workflows¶

The following GitHub workflow files are generated by pyprefab. These are located in the .github/workflow directory:

ci.yaml: runs tests against multiple versions of python, checks test coverage, and runs a linter check using ruff
publish-pypi.yaml: publishes the package to test PyPi and to PyPI (setup required, see comments in the file)

Tip

GitHub reposistories have actions enabled by default, so review the workflow files to ensure that their triggers match your expectations.

Dependabot config¶

The .github directory also contains a configuration file for GitHub Dependabot: dependabot.yaml.

This file instructs Dependabot to check your project’s Python and GitHub action dependencies each week and submit pull requests if there are any updates.

It’s important to review and merge these pull requests regularly to keep your code base secure.