Tamás Nepusz

Results 691 comments of Tamás Nepusz

Then we can probably just update the helper to expect the full filename instead and then we can have C++ tests.

If we decide to mess with the sorting algorithm, shouldn't we consider using [`timsort`][1] or some other alternative sort algorithm that has been being used in other programming languages? (timsort...

Here's a list of all the functions that still have an `igraph_bool_t loops` argument, along with my assessment of the situation. **Functions where `igraph_loops_t loops` makes sense** - [x] `igraph_adjlist_init_complementer()`...

> We already have a function to get the sparse adjacency matrix: `igraph_get_adjacency_sparse()`. And we also have `igraph_sparsemat()`, which is essentially equivalent to calling `igraph_get_adjacency_sparse()` with a null weights argument,...

Sounds good to me, feel free to go ahead and update `functions.yaml`. Ideally, the code generator for the R interface should then use the argument names from `functions.yaml` directly, and...

@szhorvat > Are you okay with this? yes, that's fine, go ahead.

Looking at the two proposed solutions, I'm definitely in favour of option 2 instead of option 1. But, I was wondering whether this closes the door for other directions, like...

The Python interface could also start using this feature if we want to achieve zero-copy data transfer between the edge list returned from `igraph_get_edgelist()` and a NumPy array, depending on...