Re: Johnsons BGL implementation for undirected graphs correct?