Graphpy

A Python implementation of edges, vertices, and graphs

Use

There are two types of each object: Undirected and Directed.

To begin, import one of the graph classes:

from graphpy.graph import UndirectedGraph

and create a graph from a dictionary of vertex vals:

# graph with vertices 'v0' and 'v1', with an edge between them

g = UndirectedGraph.from_dict({'v0': [('v1',)],
                               'v1': []})

or from a list of vertices and a list of edges:

g = UndirectedGraph.from_lists([('v0',), ('v1',)],
                               [('v0', 'v1')])

You can also initialize a graph, then add vertices and edges:

g = UndirectedGraph()

g.add_vertex('v0')
g.add_vertex('v1')
g.add_edge(('v0', 'v1'))

A vertex's val can be any hashable object, like a string, int, tuple, etc.:

# graph with vertices 'v0', 1, and (2, 2), with some edges

g = UndirectedGraph.from_dict({'v0': [(1,)],
                               1: [('v0',), ((2, 2),)],
                               (2, 2): [(1,)]})

Retrieve vertex and edge objects:

# v is an UndirectedVertex object, and e is an UndirectedEdge object

v = g.get_vertex('v0')
print v.degree

e = g.get_edge(('v0', 'v1'))
print e.vertices

Iterate through a graph's vertices:

for v in g:
    print v

Perform graph algorithms, such as search:

paths = g.search(start='v0', method='depth_first')
print paths

Create graphs with vertices and edges that have whatever attributes you want (for example, edge weights):

g = UndirectedGraph.from_lists([('v0', {'city': 'Paris'}), ('v1', {'city': 'London'})],
                               [('v0', 'v1', {'weight': 5})])

From there, use graphs to model situations, implement more graph algorithms, and whatever else you desire. And, as always, have fun!

(The tests found on Github at https://github.com/tscizzle/graphpy/tree/master/tests give many more examples and showcase the rest of the library's functionality.)

Documentation

Find the full documentation at: http://graphpy.readthedocs.org/en/latest

Installation

If you don't have pip, get pip at: https://pip.pypa.io/en/stable/installing

Run the command pip install graphpy in your terminal to get the graphpy library.

To test your installation, start a Python interpreter with the python command in your terminal and make sure you can run import graphpy in it without getting an error.

Contribute

Find the code on Github at: https://github.com/tscizzle/graphpy

Support

Contact me (Tyler Singer-Clark) at tscizzle@gmail.com with any questions or concerns.

License

The project is licensed under the MIT license.