..  -*- coding: utf-8 -*-

PyGraphviz News
===============

pygraphviz-0.36
---------------
Release date: 13 January 2008

See: https://networkx.lanl.gov/timeline

 - Automatic handling of types on init of AGraph(data): data can be
   a filename, string in dot format, dictionary-of-dictionaries,
   or a SWIG AGraph pointer.
 - Add interface to graphviz programs acyclic and tred
 - Refactor process handling to allow easier access to graphviz layout
   and graph processing programs
 - to_string() and from_string() methods 
 - Handle multple anonymous edges correctly
 - Attribute handling on add_node, add_edge and init of AGraph.
   So you can e.g. A=AGraph(ranksep='0.1'); A.add_node('a',color='red')
   A.add_edge('a','b',color='blue')


pygraphviz-0.35
---------------
Release date: 22 July 2007

See: https://networkx.lanl.gov/timeline

 - Rebuilt SWIG wrappers - works correctly now on 64 bit machines/python2.5
 - Implement Graphviz subgraph functionality
 - Better error reporting when attempting to set attributes, avoid 
   segfault when using None 
 - pkg-config handling now works in more configurations (hopefully all) 
 

pygraphviz-0.34
---------------
Release date: 11 April 2007

See: https://networkx.lanl.gov/timeline

 - run "python setup_egg.py test" for tests if you have setuptools
 - added tests for layout code
 - use pkg-config for finding graphviz (dotneato-config still works
   for older graphviz versions)
 - use threads and temporary files for multiplatform nonblocking IO
 - django example

pygraphviz-0.33
---------------
  - Workaround for "nop" bug in graphviz-2.8, improved packaging,
    updated swig wrapper, better error handling.

pygraphviz-0.32
---------------

   The release pygraphviz-0.32 is the second rewrite of the original project.
   It has improved attribute handling and drawing capabilities.
   It is not backward compatible with earlier versions.
   Earlier versions will always be available at the download site.

   This version now inter-operates with many of the NetworkX
   algorithms and graph generators.  See 
   https://networkx.lanl.gov/browser/networkx/trunk/doc/examples/pygraphviz_simple.py