Generates Homebrew resource stanzas for Python packages.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.



Build Status Code Health PyPI page MIT license

Invoked like poet foo for some package foo which is presently installed in sys.path, determines which packages foo and its dependents depend on, downloads them from pypi and computes their checksums, and spits out Homebrew resource stanzas.

poet -f foo will give you a complete Homebrew formula.

poet -s foo will write a resource stanza for a single package foo, which does not need to be installed, without considering its dependencies.

poet will use the versions of the packages that you presently have installed. If a package it wants to reference is not installed, the latest version on pypi will be downloaded and checksummed and its dependencies will not be considered.

The easiest way to use poet is to create a virtualenv, use pip or setuptools to install the target package and its dependencies in the virtualenv, and then pip install homebrew-pypi-poet and run poet inside the virtualenv.

Usage is like:

usage: poet [-h] [--single package [package ...] | --formula package |
                  --resources package]

Generate Homebrew resource stanzas for pypi packages and their dependencies.

optional arguments:
  -h, --help            show this help message and exit
  --single package [package ...], -s package [package ...]
                        Generate a resource stanza for one or more packages,
                        without considering dependencies.
  --formula package, -f package
                        Generate a complete formula for a pypi package with
                        its recursive pypi dependencies as resources.
  --resources package, -r package
                        Generate resource stanzas for a package and its
                        recursive dependencies (default).


homebrew-pypi-poet is offered under the MIT license.


homebrew-pypi-poet is maintained by Tim D. Smith. Robson Peixoto, Alessio Bogon, Julien Maupetit, and Zhiming Wang are thanked for their helpful contributions!