This snippet defines 3 functions to create in a few seconds an image of
a graph containing arbitrary python objects. It uses the module pygraphviz
(available at http://pypi.python.org/pypi/pygraphviz/).
See the example use case at the end of the module.
# module fastgraph
# created by Gribouillis for the python forum at www.daniweb.com
# November 9, 2010
# Licence: public domain
"""This module defines 3 functions (node, edge and graph) to help
create a graph (a pygraphviz.AGraph instance) linking arbitrary
python objects.
This graph can be saved in various image formats, and in dot format.
"""
import pygraphviz
def node(x):
"helper to create graphs"
return (x,)
def edge(x, y):
"helper to create graphs"
return (x, y)
def graph(items, tolabel, **kwd):
"""Create a pygraphviz AGraph instance
@ items - a sequence of node(obj), or edge(obj, obj) items (obj can be any python object)
@ tolabel - a function tolabel(obj) -> str which converts an object to string
@ **kwd - additional keyword arguments for AGraph.__init__
"""
names = dict()
the_graph = pygraphviz.AGraph(**kwd)
for uple in (tuple(t) for t in items):
for obj in uple:
if not obj in names:
names[obj] = "n%d" % len(names)
the_graph.add_node(names[obj], label=tolabel(obj), shape="box")
if len(uple) == 2:
the_graph.add_edge(*(names[obj] for obj in uple))
return the_graph
if __name__ == "__main__":
def my_items():
"""Exemple generator of graph items. Our graph contains string here"""
for x in "abcde":
yield node(x)
for s in "ab ac bd ea da".split():
x, y = iter(s)
yield edge(x, y)
def my_label(x):
"Create a label for graph objects"
return x.upper()
g = graph(my_items(), my_label)
g.draw('mygraph.png', prog='circo') # prog can be neato|dot|twopi|circo|fdp|nop
Consider posting this in the tutorials forum also so we can find it later
http://www.daniweb.com/tutorials/forum114.html
We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, learning, and sharing knowledge.