Package com.jgalgo.adapt.guava

Class GuavaNetworkAdapter<V,E>

java.lang.Object
com.google.common.graph.AbstractNetwork<V,E>
com.jgalgo.adapt.guava.GuavaNetworkAdapter<V,E>
Type Parameters:
V - the vertices type
E - the edges type
All Implemented Interfaces:
Network<V,E>, PredecessorsFunction<V>, SuccessorsFunction<V>
Direct Known Subclasses:
GuavaMutableNetworkAdapter
public class GuavaNetworkAdapter<V,E> extends AbstractNetwork<V,E>
An adapter from a JGAlgo graph to a Guava network.

The adapter is constructed with a JGAlgo graph and implements the Guava network interface, and can be used with any Guava algorithm. The adapter is a live view, so any change in the JGAlgo graph is reflected in the Guava network and vice versa, but the underlying JGAlgo graph should not be modified directly.

The Guava network is immutable, so as this adapter. A subclass of this adapter, GuavaMutableNetworkAdapter, is a mutable adapter that implements the Guava mutable network interface.

The capabilities of the Guava network, such as parallel and self edges or whether the network is directed or undirected, are determined by the underlying JGAlgo graph.

Guava support different element orders for the nodes and the incident edges. This adapter uses the default element order, which is unordered, and cannot be changed as the underlying JGAlgo graph does not support any other order.

Among Guava Graph, ValueGraph and Network, the network is the most similar to JGAlgo graphs, as vertices and edges have unique identifiers, and queries of edges are answered with the edges identifiers. On the other had, the Graph and ValueGraph do not support unique identifiers for the edges, and operations on edges are addressed by a pair of nodes. The ValueGraph does associate a value with each edge, be it does not have to be unique, and it is more similar to weights in JGAlgo graphs.

For adapting the other way around, from Guava Network to JGAlgo, see GuavaNetworkWrapper.

Author:
Barak Ugav
See Also: