Admin message

Please use the issue tracker only to report bugs (i.e. errors in the library that need to be fixed) or feature requests.

Please check if the current git version also has the same problem. An easy way to do this is to use google colab!

For questions about how to compile, install or use the library, please use instead the web forum at https://forum.skewed.de/c/graph-tool.


(If unsure, use the forum first.)


IMPORTANT: When opening new issues, please fill out the required information listed in the issue template!

Make the installation easier?

As a researcher in software ecosystems, I recently encountered performance issues on (quite) large graphs with NetworkX. I then considered using graph-tool. However, as a Fedora user, I had great difficulty in finding an easy solution to install graph-tool. I didn't find any package targeting Fedora/CentOS, I didn't manage to turn the provided .deb into a valid .rpm, and I didn't want to spend too much time compiling graph-tool and its dependencies.

I then moved on to igraph and its Python binding which were fine for my purposes, but I felt that I should give it a try with graph-tool. I think you should consider making the installation easier for (some) Linux users. The first solution is to provide packages targeting Fedora/CentOS, which is quite a major distribution nowadays. The second solution is to provide wheels for (major) architectures on PyPI, which can be quite a difficult (and annoying) task. The third one, which could be easier for you, is to consider Conda and, more specifically, the open-source Conda-Forge (see https://conda-forge.github.io) that provides a very simple mechanism for developers to have their package automatically build against the three major operating systems.

Btw, thanks for your job. I hadn't a chance yet to test graph-tool, but I enjoyed reading the documentation and I hope I'll be able to perform some research with it (I've to admit I really don't like the API of igraph's python binding nor its documentation).

Assignee Loading
Time tracking Loading