I have a small project that requires a third-party module that's rarely updated (once every other year or so) and not maintained by any package management system. The context, here, is cognitive science & research, the third party module is for linking proprietary hardware to custom scripting, in the house project, potentially, may end up seeing wider (eventually open source) use within cog. psych so I want to plan for distribution.
Can anyone poke holes in or thumbs up this approach:
#!/usr/bin/env python
from distutils.core import setup
try:
import third_party_package
install_packages = ['in_house_package']
except ImportError:
print "Warning: 'third_party_package' not found in PYTHON_PATH, installing version in_house_package v. <VERSION>"
install_packages = ['in_house_package', 'in_house_package.third_party_package']
setup(
name='InHousePackage',
version = '0.1',
description = 'A framework for building psychological experiments in Python',
author = 'me',
author_email = 'my_address',
url = 'a local git',
packages=install_packages,
requires = ['numpy']
)
third_party_package
is critical and will, almost certainly, never be updated to Python 3. – Jonline Dec 15 '14 at 13:42