Tamás Nepusz

> I ran with the example in the OP, there the p-value comes out as 0.98 I can get 0.98 only with the current implementation of `plfit` if I use...

Seconds or even more, compared to milliseconds with the previous version (which used an approximation that was not even scientifically valid).

TODO list: - [ ] `igraph_pagerank_algo_t` should be extended with an "automatic" member - [ ] `igraph_barabasi_algorithm_t `should be extended with an "automatic" member - [ ] `igraph_fas_algorithm_t` should be...

> Do we want to standardize the naming of such parameters? There's really no need to. We can invent as many "standards" as we want but even we are not...

Fixed in 797f0a9. There does not seem to be a consensus on the position of the `weights` argument in shortest path related functions; I added it to the second position...

> Do you object to my adding this parameter? No, feel free to go ahead. The more specific functions could be removed if we are absolutely sure that these functions...

As part of this issue I'll rename all functions that have a BLISS-specific interface to have the suffix `_bliss` in their names, and provide an algorithm-agnostic variant under the original...

I decided not to add an `igraph_isomorphism_method_t` argument. If the user wants to specify exactly which method to call, they are better off calling the lower-level function directly.

@szhorvat What about `igraph_get_isomorphisms_vf2()` and `igraph_get_isomorphisms_vf2_callback()`? I believe it makes sense to keep them as is, with the current name, as the user may provide two graphs that are isomorphic,...

Actually, maybe even keeping `igraph_count_isomorphisms_vf2()` makes sense. The user might want to ask the question: "Are these two graphs isomorphic and if so, what is the automorphism count?". This can...