Python wrapper library for udunits2
cf_units instead.Note: this package is no longer being developed or supported. We recommend the use of
#Prerequisites These steps are tailored for a Mac OSX environment. The library should function on linux as well, but modifications to the following setup may be required
/usr/bin/ruby -e "$(curl -fsSL https://raw.github.com/gist/323731)"
Install git, python, and udunits with Homebrew
brew install git python udunits
##Optional Python Dependencies These dependencies are optional, but tend to make life a little easier.
Virtualenv allows creation of isolated Python environments. See more here.
Install virtualenv and virtualenvwrapper modules for your python 2.7 installation Note: This may require OSX's XCode (can use XCode 3.3 free version)
easy_install --upgrade virtualenv easy_install --upgrade virtualenvwrapper
Setup a virtualenv to run coverage-model (use any name you like):
mkvirtualenv --python=python2.7 udunitspy
Ensure you're 'working in' the virtualenv you just made:
You can just use easy_install if you prefer.
#ipython ipython is an 'extended' python shell which provides useful functionality such as tab-complete, history, etc. If you wish to use it, you can install it by running:
pip install ipython
##Required Python Dependencies These dependencies are required for the library to function
#Numpy The numpy library is finiky about installing as a dependency via setuptools, so ensure you have it installed in your virtualenv by running:
pip install numpy
#Installation as a library Ensure your environment matches the prerequisites prior to running the steps below
###As a local library Run the following command:
pip install udunitspy
###As a project dependency The library can be included in your project by adding the following to your setup.py:
install_requires = [ ..., udunitspy, ],
Once the library is installed, you should be able to access and use it from the ipython/python shell:
In : from udunitspy import Unit In : m = Unit('m') In : ft = Unit('ft') In : m.are_convertible(ft) Out: True In : m2ft = m.get_converter(ft) In : m2ft.evaluate([20, 22.3, 25.6]) Out: array([ 65.6168 , 73.162732, 83.989504])
#Unit Tests Unit tests can be run with the following command:
#Source installation (for development) Clone the git repository:
git clone https://github.com/blazetopher/udunitspy.git cd udunitspy
###Install Install the uduintspy library for development:
python setup.py develop
During development, there is no need to reinstall. Changes to the source are automatically reflected in your working environment.
#Uninstall NOTE: Please ensure these libraries aren't needed by other software before uninstalling!!
The library and dependencies can be uninstalled by calling:
pip uninstall -y numexpr pytest py pytest-cov coverage cov-core udunitspy
Leftover files & directories can be removed by running: