GroteGnoom

Maybe we could use NaN/Inf weights for that? Your proposal is indeed something we want, but we don't have a proper system for it yet: https://github.com/igraph/igraph/issues/2222

Note: what's required for direct betweenness is cutoff versions of `igraph_get_all_shortest_paths(_dijkstra)()`

> This brings up an interesting question: > > Should the comparison with the cutoff be done with tolerances, even in the current functions? Do we get paths which are...

> In algorithms where each weight value is accessed at most once, such as A*, the weights could be checked at access time instead of at the beginning of the...

I want to work on this one, should I add a 'directed' parameter to ignore edge directions?

> There is a related task that I'm not sure if I created an issue for, and you may be interested in: `rewire_directed_edges` should support avoiding multi-edges. The same requirement...

> The second branch does `(long int)(type->get(rng->state) / ((double)max + 1) * (h - l + 1) + l);`, which looks fine to me. This gets you all the right...

Slightly off topic: The doc is incorrect, it says `* \function igraph_adjlist_init` instead of `* \function igraph_adjlist_init_from_inclist`