Tulip is an open source, cross-platform, data visualization framework mainly dedicated to the analysis, the drawing and the visualization of very large graphs (up to the million of nodes and edges). It has been created by David Auber, from LaBRI (Laboratoire Bordelais de Recherche en Informatique) and University of Bordeaux, and maintained over the years by an average size development team. Until the 5.1 release, the main Tulip code repository (subversion based) was hosted on SourceForge. That repository is now in read-only mode and future development of Tulip will be hosted here.
Tulip is an information visualization framework dedicated to the analysis and visualization of relational data. Tulip aims to provide the developer with a complete library, supporting the design of interactive information visualization applications for relational data that can be tailored to the problems he or she is addressing.
Written in C++ the framework enables the development of algorithms, visual encoding, interaction techniques, data models, and domain-specific visualizations. One of the goal of Tulip is to facilitates the reuse of components and allows the developers to focus on programming their application. This development pipeline makes the framework efficient for research prototyping as well as the development of end-user applications.
The Tulip framework offers numerous features, notably:
An efficient graph data model in terms of memory usage for storing large networks and the attributes of their elements (called properties in the Tulip semantics). It is also one of the few that offer the possibility to efficiently define and navigate graph hierarchies or cluster trees (nested subgraphs).
Several graph file formats for serializing such a model to disk, notably the TLP format based on a Lisp syntax for easy parsing but also the TLP binary format for faster graph saving and loading.
A large variety of graphs algorithms: clustering, metric, layout ... As Tulip is dedicated to graph visualization, it is provided with numerous state of the art graph layout algorithms but also a bridge to the Open Graph Drawing Framework.
A hardware accelerated graph rendering engine written in OpenGL, highly customizable in terms of visual encoding for graph nodes and edges, in order to efficiently generate aesthetic and interactive visualizations.
Multiple visualization components (called views in the Tulip semantics) for analyzing graph data using other representations than the classical node link diagram one: matrix, histograms, scatter plots, parallel coordinates, ...
Python bindings for the main Tulip C++ API, giving to Tulip scripting facilities for manipulating graphs loaded from its main graphical user interface. The bindings can also be obtained from the Python Packaging Index.
A plugin based architecture for easily extend the capability of the framework with new graph import mechanisms, graph algorithms, visualization components, ... Tulip plugins can be written in C++ or Python.
A graphical user interface, based on the Qt framework, enabling to easily interact and manipulate the different components of the framework.
Below are some links to relevant documentation resources about the Tulip framework:
Tulip is a cross-platform framework and can be compiled or installed on FreeBSD, main Linux distributions (Debian, Fedora, OpenSUSE or Ubuntu, the packages provided are not necessarily in synch with latest Tulip version), MacOS and Windows.
Compiling from scratch
Tulip can be easily compiled on every supported platforms. However, that process can take some times depending on your system configuration.
The following dependencies are required to build Tulip:
- CMake >= 2.8.12
- A C++11 compiler : GCC >= 4.8.1, Clang >= 3.3 or Microsoft Visual Studio >= 2013
- Qt >= 5.5.0
- OpenGL >= 2.0
- GLEW >= 1.4
In order to build the Python components, the following dependencies are needed:
- Python >= 2.6
- SIP >= 4.19.14 (if SIP can not be found or its version does not match the required one, it will be compiled using a copy of its source code in the Tulip tree)
The following dependencies are also needed but they will be compiled from the Tulip source tree if they can not be found on your system:
In order to generate the documentation, the following tools must be installed:
- Sphinx to build the User Manual, Developer Handbook and Python bindings documentation
- Doxygen to build the C++ API documentation
If you are a Linux user, all these dependencies can be installed with the package manager of your distribution.
If you are a Windows user, we recommend to use MSYS2 as it greatly facilitates the build of Tulip on that platform (notably by providing up to date compilers and precompiled dependencies).
Below are some scientific publications related to Tulip (see Tulip bibliography for more information):
David Auber, Daniel Archambault, Romain Bourqui, Maylis Delest, Jonathan Dubois, Antoine Lambert, Patrick Mary, Morgan Mathiaut, Guy Mélançon, Bruno Pinaud, Benjamin Renoust and Jason Vallet. TULIP 5. In Encyclopedia of Social Network Analysis and Mining, Springer, pages 1-28, 2017.
David Auber, Romain Bourqui, Maylis Delest, Antoine Lambert, Patrick Mary, Guy Mélançon, Bruno Pinaud, Benjamin Renoust and Jason Vallet. TULIP 4. Research report. LaBRI - Laboratoire Bordelais de Recherche en Informatique. 2016.
David Auber, Daniel Archambault, Romain Bourqui, Antoine Lambert, Morgan Mathiaut, Patrick Mary, Maylis Delest, Jonathan Dubois, and Guy Mélançon. The Tulip 3 Framework: A Scalable Software Library for Information Visualization Applications Based on Relational Data. Technical report RR-7860, INRIA, January 2012.
Antoine Lambert and David Auber. Graph analysis and visualization with Tulip-Python. EuroSciPy 2012 - 5th European meeting on Python in Science, Bruxelles.
David Auber. Tulip - A Huge Graph Visualization Framework. Graph Drawing Software, Springer, pages 105-126. 2004.
Tulip is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Tulip is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.