rootpy aims to provide a more feature-rich and pythonic interface with the ROOT libraries on top of the existing PyROOT interface.

More specifically, rootpy provides:

• an interface between ROOT and matplotlib. Don't like the way your plots look in ROOT? Simply use rootpy to plot your ROOT histograms or graphs with matplotlib instead
• easier manipulation of histograms, graphs, and TVector/TLorentzVectors. rootpy provides classes that inherit from these ROOT classes and implement the Python arithmetic operators
• an easy way to create and read ROOT TTrees and a mechanism for defining objects and collections of objects whose attributes are TTree branches. You may also decorate TTree objects with additional methods and attributes. See examples/tree.
• easy navigation through TFiles. rootpy wraps TFile and implements the natural naming convention so that objects may be retrieved with myFile.someDirectory.treeName, for example
• the ability to convert ROOT TFiles containing TTrees into HDF5 format with PyTables
• a framework for parallelizing processes that run over many TTrees
• roosh, a Bash-like shell environment for the ROOT TFile
• a collection of useful command line scripts: root-ls, root-cp, root-tree-ls, and others.

At least Python version 2.6 and ROOT with PyROOT enabled. matplotlib, numpy, PyTables, and PyYAML are optional.

The easiest way to install rootpy is with pip. To install for all users:

sudo pip install rootpy

To install in your home directory:

pip install --user rootpy

If you have obtained a copy of rootpy yourself use the setup.py script to install. To install for all users:

sudo python setup.py install

To install in your home directory:

python setup.py install --user

To install the optional requirements (first download a source distribution if you haven't already):

pip install -U -r optional-requirements.txt

Clone the code from github.com with git:

git clone git://github.com/ndawe/rootpy.git

or checkout with svn:

svn checkout http://svn.github.com/ndawe/rootpy

see examples/*

Please contact me (Noel dot Dawe AT cern dot ch) if you have ideas or contributions. And of course feel free to fork rootpy at GitHub.com and later submit a pull request.

rootpy needs attention in these areas:

• Documentation
• Tutorials
• A website displaying the above (currently here: http://ndawe.github.com/rootpy)
• Unit testing
• Brenchmarking performance (i.e. Tree read/write)
• Finishing the server/worker code for distributed computing across multiple nodes
• Creation of a TBrowser alternative using PyGTK
• Creation of a framework for managing datasets (using SQLite as a back-end? with revision control?)
• Additional features anyone would like to implement