Etienne dg

@Tokazama Have you a more precise idea of your suggestion ? I'm not sure to understand what it would looks like.

I agree that returning the created vertex would be great, but that would be a big breaking change... Maybe we can consider it for a 2.0 release ?

A function which more generally returns the induced subgraph of a set of vertices could be useful. It could also be a good idea to define views on graphs where...

I also think it's good to have. Regarding your comment on edges, I also think it would be good to have a distinction between directed on non-directed edges, but we...

The old implementation was very wrong and failing also for unweighted graphs (see the example provided in the [original issue](https://github.com/sbromberger/LightGraphs.jl/issues/1589)). It is expected that the new implementation is much slower...

Just realized that Stoer-Wagner does not work on directed graphs, reverted the commit and disabled the function for directed graphs.

Hmm, I'm not sure what happened here? Did you try rebasing the PR?

new PR [here](https://github.com/JuliaGraphs/Graphs.jl/pull/168)

I'm against throwing away these algorithms: - This might break existing code, this is much harder to reimplement the algorithm using JuMP than just swapping GraphFlows for GraphsOptim. - This...

I also think we should bring the standard flow algorithms to Graphs.jl. For the moment, we could keep the ones that use LP solvers on GraphOptim, and bring them to...