Tamás Nepusz

Adding this to 0.10.0 for the time being as I don't see yet how the stimulus-based rewrite will tie into the `python-igraph` repo. It might end up being such a...

I'm not thinking about completely doing away with a middle layer that provides "pythonic" access to igraph's core functions, I'm just trying to minimize the amount of code we need...

Short enumeration: * `implementation` is used by `Graph.Barabasi`, `Graph.pagerank`, `Graph.personalized_pagerank`, `Graph.induced_subgraph` and `Graph.community_spinglass` * `algorithm` is used by `Graph.eigen_adjacency` * `method` is used by `Graph.Degree_Sequence`, `Graph.Realize_Degree_Sequence`, `Graph.Tree_Game`, `Graph.feedback_arc_set`, `power_law_fit` and...

This is not a bug; the documentation states that when `weighted=NULL` (which is the default), igraph will create an unweighted graph and the _number_ of edges is specified by the...

Thanks for the heads up; I've turned on `warnPartialMatchArgs` on my dev machine and I'll fix stuff as I encounter them. The biggest offenders seem to be `rep(..., length=...)` and...

@iosonofabio If it is too complicated to do with Matplotlib, let me know and we can simply say that this is not supported in Matplotlib or Plotly.

It's amazing what you can do in the browser nowadays. I don't have any experience with these tools, but it seems quite sure that you also need to convert the...

The issues about `igraph_*_yylex_destroy()` are probably easy to fix. We declare these functions on our own in our C files with a `void` return type, while it seems like `bison`...

https://github.com/igraph/igraph/commit/c307fc4089435f95587df7394a56a0f443a804f1 changes the `yylex_destroy()` functions to return `int`; this should fix most of the problems above. The remaining two are about `s_cat` and `s_copy`; I need to test those more...

Flex generates `int` as the return type, although I have no idea why - a well-behaved "destroy" function shouldn't fail. `IGRAPH_FINALLY()` is okay with such a function, it will simply...