The FordFulkerson class represents a data type for computing a Maximum st-flow and Minimum st-cut in a flow network.

This implementation uses the Ford-Fulkerson algorithm with the Shortest augmenting path heuristic. The constructor takes time proportional to E V (E + V) in the worst case and extra space (not including the network) proportional to V, where V is the number of vertices and E is the number of edges. In practice, the algorithm will run much faster. Afterwards, the inCut() and value() methods take constant time.

If the capacities and initial flow values are all integers, then this implementation guarantees to compute an integer-valued maximum flow. If the capacities and floating-point numbers, then floating-point roundoff error can accumulate.