jupyter-packaging

What it is and what it does

Jupyter Packaging provides build-time utilities for Python packages that combine Python code with JavaScript assets or other pre-build steps. It wraps setuptools to inject custom build commands (like npm builds) before package distribution or development installation. The package is designed to be used as a build requirement in pyproject.toml, not as a runtime dependency—it helps package authors handle the complexity of building hybrid Python/JavaScript projects without burdening end users.

The package offers three integration patterns: as a build requirement with custom setup.py commands, as a full build backend via jupyter_packaging.build_api, or as a vendored setupbase.py module. It handles common Jupyter packaging tasks like managing pre-build steps, ensuring generated assets are included in distributions, and supporting both wheel and sdist formats through modern build tools.

Use it for:

• Build a Jupyter extension that includes TypeScript/JavaScript components compiled during package installation.
• Manage npm build steps in a Python package without requiring end users to have Node.js installed.
• Ensure generated assets (CSS, bundled JS) are correctly included in wheel distributions and source tarballs.
• Develop a Jupyter lab plugin with a Python backend and npm-managed frontend build pipeline.
• Configure data files and assets for Jupyter packages using setuptools with pre-build hooks.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides build tools and utilities for packaging Jupyter Python projects that require pre-build steps, including JavaScript compilation and asset management.

Yes, if you are packaging a Jupyter project with JavaScript or other pre-build requirements. Install as a build-only dependency in pyproject.toml, not as a runtime requirement. The package is actively maintained, has no known vulnerabilities, and is widely used in the Jupyter ecosystem. Do not add it to your package's runtime dependencies.

Install

jupyter-packaging on PyPI

pip install jupyter-packaging

uv add jupyter-packaging

poetry add jupyter-packaging

Installing jupyter-packaging

Before you install

Low install friction with a pure-Python wheel distribution. Actively maintained with recent commits; intended as a build-time dependency only, not a runtime requirement for end users.

License in practice

BSD 3-Clause License permits commercial and private use with minimal restrictions; you may redistribute and modify freely provided you retain the license notice and disclaimer.

Quickstart

# In pyproject.toml
[build-system]
requires = ["jupyter_packaging>=0.10,<2"]
build-backend = "setuptools.build_meta"

# In setup.py
from jupyter_packaging import wrap_installers, npm_builder
builder = npm_builder()
cmdclass = wrap_installers(pre_develop=builder, pre_dist=builder)
from setuptools import setup
setup(cmdclass=cmdclass)

Intended for use during package build/development only; should not be listed as a runtime dependency in your package's requirements.

Verify before relying

• Whether the package supports Python 3.11+ despite classifiers listing only up to 3.10.
• Current status of the npm_builder for non-Node.js JavaScript toolchains or alternative build systems.
• Whether the package works with modern build frontends beyond setuptools.

Package facts

Evidence: jupyter_packaging-0.12.3-py3-none-any.whl

Keywords: ipython, jupyter, packaging

Tags

More Build Tools packages