GroteGnoom

In igraph_i_barabasi_game_bag, `new_edges` is calculated by taking a `igraph_vector_int_sum`, which sums the degrees of vertices. Should we make an overflow-safe `igraph_vector_int_sum` ? It seems slow, but the slowdown should not...

If a user supplies a number of edges, should it be checked that its not greater than IGRAPH_ECOUNT_MAX?

I thought this would happen because of lots of little mallocs. I'll try to get some profiling going later 🙂

It happens because igraph_cattribute setting and getting functions use: ``` igraph_i_cattributes_t *attr = graph->attr; igraph_vector_ptr_t *val = &attr->val; igraph_bool_t l = igraph_i_cattribute_find(val, name, &j); ``` Where attr is NULL, and...

> IMO this doesn't necessarily need to be fixed. It is the nature of C programming that when some preconditions are not satisfied when calling a function, the result is...

I would like to work on this. I looked at `igraph_st_mincut`, which only calls `igraph_maxflow`, which is already tested via `igraph_maxflow_value`. But the resulting cut is not actually checked there....

I made a wiki addition about adding tests here: https://github.com/GroteGnoom/igraph/wiki/Adding-tests But you can't really do wiki pull requests, and I also can't add new topics to the `discourse`. What's the...

igraph_write_graph_lgl can be crossed off the list because of #1601

#1622 fixes rewire_directed_edges #1625 fixes lastcit_game

I am going to work on the communities, starting with igraph_compare_communities. Just for some variation :slightly_smiling_face: