Tamás Nepusz comments

Results 691 comments of


                                            Tamás Nepusz

Update igraph-scg for igraph 0.10

Yes, in the meanwhile I gave up on 1 and went for an explicit `weights` argument. The argument list is going to be ugly for all functions, plus we need...

Update igraph-scg for igraph 0.10

I'm getting all sorts of errors and test failures; I'm going to return to this tomorrow.

Update igraph-scg for igraph 0.10

`igraph-scg` updated again and now it compiles and passes all tests.

Update igraph-scg for igraph 0.10

Just checked it again and it still compiles with the most recent `master` revision ( + all unit tests pass).

Update igraph-scg for igraph 0.10

Updated for the soon-to-be igraph 0.10.0-rc.2.

Update igraph-scg for igraph 0.10

Still compiles correctly with revision 540d5009e (head revision of the `develop` branch of igraph as of today)

degree() and strength() should have fine-grained loop control

`igraph_degree()` and `igraph_strength()` currently has a `loops` argument, which is a boolean, and a `mode` argument, which can be `IGRAPH_IN`, `IGRAPH_OUT` or `IGRAPH_ALL`. **When loops is false**: This is easy,...

degree() and strength() should have fine-grained loop control

As an experiment, I went ahead with my proposal in the `feat/loop-edge-handling` branch to see what problems it may cause.

Check functions for integer overflow

I guess so as its an O(1) check and we should probably have an `IGRAPH_CHECK_EDGE_COUNT()` macro anyway, but I'd leave it up to @szhorvat to decide.

Check functions for integer overflow

Also, I wonder whether we should also check the "number of vertices" argument supplied by the user in graph constructor functions. Sure, we could leave it up to `igraph_create()` or...