Szabolcs Horvát
Szabolcs Horvát
Decisions to be made: 1. Do we guarantee any sort of behaviour at all when some entries are not integers? It is tempting not to, as it simplifies things and...
My suggestions: 1. Guarantee a behaviour, but prioritize integer values. I.e. don't try to implement a "nice" behaviour for non-integer. 2. No strong opinion, leaning towards a warning (or not...
I did this work on this before, see #2417. I'll have look check and remind myself about what exactly was done already and what is still missing.
@ntamas The interface here is based on the stub function that was accidentally left over in 0.10 (though not exposed). As I said in chat, I am not sure that...
We didn't forget this PR, just need to find the time to go over it. It's a non-trivial amount of work. The `m[v] = u` approach is fine internally, but...
I had a quick look while testing Copilot, but unfortunately I still don't have the capacity to review this PR in full at this moment. I hope that you are...
What would be a good solution here? Perhaps skip standard Pajek attributes that contain NaN and issue a warning? `igraph_vector_is_any_nan()` should make it easy enough. @krlmlr Is it easy to...
I wouldn't fix this before 1.0 since @ntamas already modified the Pajek reader/writer on `develop`, and it is not a high-impact bug.
I suggest replacing any NaN with Pajek's defaults for that specific attribute when writing the graph.
Alright, then let's remove this from the milestone. If needed, we can add a more flexible alternative with a new name in the future, and deprecate this one.