GraphUI

A python-based visual graph editor. It can create and edit GraphViz files specified in the DOT language.

This is a fork of the GraphUI utility from the Enough project, originally developed by Noam Lewis and Eyal Lotem, but abandoned since 2007.

This repository was automatically exported from code.google.com/p/enough using the Google Code Github Exporter.

Below are the contents of the Graphui page from the original Enough wiki:

Status

We have re-implemented the PackageLibrary - so now we need a major re-write of Graphui as well.

Download

The latest version (BEFORE the anticipated rewrite) is: Latest Version

Features

Graphui (pronounced grafoo-ee) is an attempt at a generic graph editing gui. The code is in the svn under graphui.

Short Demo video

Screenshot:

Here are the current "features":

Automatic layout using Dot, Neato & Twopi (Graphviz).
You can switch layout engines in real time
Cross-platform - tested on Linux (ubuntu) and Windows
Save and load graphs
Connect, disconnect, add and remove nodes
Zoom & pan
Stretch or keep DOT's aspect ratio
Undo/redo
Multi-line text labels for nodes and edges
Record animation - makes a series of BMP files you can later turn into a video
using videowriter or some other tool.
All changes are animated to make it easy to see what is changing (and to make it cooler!)

More coming soon!

Requirements: Python 2.5, Pygame, Twisted, and Graphviz (aka dot). On Windows you will need also pywin32.

For better performance get Pyrex too (and run setup.py build make sure to use the .so files).

Here is a short animation showing Graphui switching between layout engines:

Applications

Graphui is still under development. Here are some ideas for applications.

Diagram editor - the simplest most obvious application. Use Graphui instead of manual-layout programs (such as Visio and Dia) to easily create flow charts, state machine diagrams, etc.
Graph visualization - because Graphui's format is simply pickled Python objects, you can easily create tools that make graphs from any dataset and then view/edit the graph in Graphui.
Graph algorithms demonstration and visualization - a researcher/student may easily extend Graphui's code to work on graph algorithms in a live environment, instead of iterating between running his algorithm test program and viewing the result seperately.

Future applications

Graphui may be suitable for testing a Live programming environment. We will have to add features for viewing the code in different ways (not just as a Dot graph), especially as a compact tree (like the one used in graphical file managers to represent the directory tree).