Requirements
-------------

To use NetworkX you need Python version 2.3 or later http://www.python.org/

See below for tips on installing Python.


Installing Pre-built NetworkX packages
--------------------------------------

The following pre-built NetworkX packages are available:

Windows
~~~~~~~
Download and run the latest version of the Windows installer (.exe extension).

OSX 10.5
~~~~~~~~
Download and install the latest mpkg. 

Linux
~~~~~
Debian packages are available at http://packages.debian.org/unstable/graphics/python-networkx


Python Eggs for All Platforms 
-----------------------------

Python eggs are prepackaged distributions that can be automatically 
installed using EasyInstall http://peak.telecommunity.com/DevCenter/EasyInstall

Run

::

   easy_install networkx

and an attempt will be made to find and install an appropriate version
that matches your operating system and Python version. 

Installing from Source
----------------------

You can install from source by downloading a source archive file
(tar.gz or zip) or by checking out the files from the Subversion repository.
NetworkX is a pure Python package (you don't need a compiler to build
or install it).

Source Archive File
~~~~~~~~~~~~~~~~~~~
  - Download the source (tar.gz or zip file).
  - Unpack and change directory to networkx-x.xx where x.xx is the version number
  - Run "python setup.py install" to build and install 
  - (optional) cd networkx/tests and run "python setup_egg.py test" to execute the tests

SVN Repository
~~~~~~~~~~~~~~
Make sure you have Subversion installed (SVN)

  - Check out the networkx trunk: svn co https://networkx.lanl.gov/svn/networkx/trunk networkx
  - Run "python setup.py install" to build and install 
  - (optional) run "python setup_egg.py test" to execute the tests


Optional packages 
-----------------

NetworkX will work without the following optional packages.
But to enable all functionality you will need the following:

 - To enable drawing networks with matplotlib you need

      - numpy		 http://numpy.scipy.org/
      - Matplotlib       http://matplotlib.sourceforge.net/

 - To enable graphviz layout and drawing features you need

      - pygraphviz	 http://networkx.lanl.gov/pygraphviz/
      - Graphviz         http://graphviz.org/


Installing Optional Packages
----------------------------

For Windows and OSX (non-fink, non-MacPorts) the easiest way to
get Matplotlib/Numpy/Scipy installed is to follow the directions for
installing pre-built Scipy binary packages at http://scipy.org/Download

For Linux systems try installing binary packages through your package
management system.

Matplotlib
~~~~~~~~~~

NetworkX uses Matplotlib for drawing graphs.  
For installation see the instructions at 

 - http://matplotlib.sourceforge.net/installing.html

Matplotlib uses numpy for numerical array computing (see above comment
on how to install Matplotlib/Numpy/Scipy all together). 

Additional useful packages
~~~~~~~~~~~~~~~~~~~~~~~~~~
These are extra packages you may consider to use with NetworkX

      - IPython, interactive Python shell, http://ipython.scipy.org/
      - SciPy, scientific computing package, http://scipy.org/
      - sAsync, persistent storage with SQL, http://foss.eepatents.com/sAsync  
      - PyYAML, structured output format, http://pyyaml.org/


Installing Python
-----------------

If you don't have Python installed on your system:

Windows
~~~~~~~
The easiest way to get Python and most optional packages is to install
the Enthought Python distribution: http://code.enthought.com/enthon/

Other options are

 - Download from the official Python site at
   http://www.python.org/download/

 - ActiveState also distributes a binary version 
   http://activestate.com/Products/ActivePython/?mp=1

OSX
~~~
OSX 10.5 ships with Python version 2.5.  If you
have an older version we encourage you to download
a newer release. Pre-built Python packages are available from 

 - Pythonmac  http://www.pythonmac.org/packages/ 

Other options are:

 - ActiveState http://activestate.com/Products/ActivePython/?mp=1

 - Download from the official Python site at
   http://www.python.org/download/

If you are using Fink or MacPorts, Python is available through both
of those package systems.