[jgrapht-users] Is the target vertex of an edge stored differently than the same vertex in the vertex set?

[Steven B. <ste...@sa...>]

I seem to have come across a non-intuitive discrepancy between how vertexes
are stored in jgrapht graphs (in this case a SimpleDirectedGraph) in the
vertexSet, vs. how they are stored as the target of an edge (see below for
my usage of getEdgeTarget). I had thought they were identical - i.e. they
point to the same object (a SteveVertex in this case) and have the same
hash code - but they seem to be located in different places somehow, per
symptom below.

When my code below comes across a loop, in which it encounters the original
start vertex as the target of an edge, it does not show this vertex as
having been visited, even though the hashCode (SteveVertex@vertex1) is
identical to the first entry of the vertexSet which DOES show the vertex as
having been visited.

Any help would be greatly appreciated in diagnosing this.

private static SimpleDirectedGraph<SteveVertex, DefaultEdge> graph;
private static void SteveDepthFirstSearch(SteveVertex startVertex)
 {
     if (!startVertex.visited()) {
         startVertex.visit();
         for (DefaultEdge e : graph.outgoingEdgesOf(startVertex))
             SteveDepthFirstSearch(graph.getEdgeTarget(e));

         startVertex.setFinishOrder(counter);
         counter++;
     }

 }

public static void main(String[] args) {

    ...

    for (SteveVertex v: graph.vertexSet())
        SteveDepthFirstSearch(v);

    ...
}

-- 


This communication is confidential and subject to and governed by the 
confidentiality and use restrictions contained in Saama’s Electronic 
Communications Disclaimer. <http://www.saama.com/disclaimer>