GraphsExplained (original) (raw)
Graphs, Explained
Guava's common.graph
is a library for modelinggraph-structured data, that is, entities and the relationships between them. Examples include webpages and hyperlinks; scientists and the papers that they write; airports and the routes between them; and people and their family ties (family trees). Its purpose is to provide a common and extensible language for working with such data.
Definitions
A graph consists of a set of nodes (also called vertices) and a set ofedges (also called links, or arcs); each edge connects nodes to each other. The nodes incident to an edge are called its endpoints.
(While we introduce an interface called Graph
below, we will use "graph" (lower case "g") as a general term referring to this type of data structure. When we want to refer to a specific type in this library, we capitalize it.)
An edge is directed if it has a defined start (its source) and end (itstarget, also called its destination). Otherwise, it is undirected. Directed edges are suitable for modeling asymmetric relations ("descended from", "links to", "authored by"), while undirected edges are suitable for modeling symmetric relations ("coauthored a paper with", "distance between", "sibling of").
A graph is directed if each of its edges are directed, and undirected if each of its edges are undirected. (common.graph
does not support graphs that have both directed and undirected edges.)
Given this example:
graph.addEdge(nodeU, nodeV, edgeUV);
nodeU
andnodeV
are mutually adjacentedgeUV
is incident tonodeU
and tonodeV
(and vice versa)
If graph
is directed, then:
nodeU
is a predecessor ofnodeV
nodeV
is a successor ofnodeU
edgeUV
is an outgoing edge (or out-edge) ofnodeU
edgeUV
is an incoming edge (or in-edge) ofnodeV
nodeU
is a source ofedgeUV
nodeV
is a target ofedgeUV
If graph
is undirected, then:
nodeU
is a predecessor and a successor ofnodeV
nodeV
is a predecessor and a successor ofnodeU
edgeUV
is both an incoming and an outgoing edge ofnodeU
edgeUV
is both an incoming and an outgoing edge ofnodeV
All of these relationships are with respect to graph
.
A self-loop is an edge that connects a node to itself; equivalently, it is an edge whose endpoints are the same node. If a self-loop is directed, it is both an outgoing and incoming edge of its incident node, and its incident node is both a source and a target of the self-loop edge.
Two edges are parallel if they connect the same nodes in the same order (if any), and antiparallel if they connect the same nodes in the opposite order. (Undirected edges cannot be antiparallel.)
Given this example:
directedGraph.addEdge(nodeU, nodeV, edgeUV_a); directedGraph.addEdge(nodeU, nodeV, edgeUV_b); directedGraph.addEdge(nodeV, nodeU, edgeVU);
undirectedGraph.addEdge(nodeU, nodeV, edgeUV_a); undirectedGraph.addEdge(nodeU, nodeV, edgeUV_b); undirectedGraph.addEdge(nodeV, nodeU, edgeVU);
In directedGraph
, edgeUV_a
and edgeUV_b
are mutually parallel, and each is antiparallel with edgeVU
.
In undirectedGraph
, each of edgeUV_a
, edgeUV_b
, and edgeVU
is mutually parallel with the other two.
Capabilities
common.graph
is focused on providing interfaces and classes to support working with graphs. It does not provide functionality such as I/O or visualization support, and it has a very limited selection of utilities. See the FAQfor more on this topic.
As a whole, common.graph
supports graphs of the following varieties:
- directed graphs
- undirected graphs
- nodes and/or edges with associated values (weights, labels, etc.)
- graphs that do/don't allow self-loops
- graphs that do/don't allow parallel edges (graphs with parallel edges are sometimes called multigraphs)
- graphs whose nodes/edges are insertion-ordered, sorted, or unordered
The kinds of graphs supported by a particular common.graph
type are specified in its Javadoc. The kinds of graphs supported by the built-in implementations of each graph type are specified in the Javadoc for its associated Builder
type. Specific implementations of the types in this library (especially third-party implementations) are not required to support all of these varieties, and may support others in addition.
The library is agnostic as to the choice of underlying data structures: relationships can be stored as matrices, adjacency lists, adjacency maps, etc. depending on what use cases the implementor wants to optimize for.
common.graph
does not (at this time) include explicit support for the following graph variants, although they can be modeled using the existing types:
- trees, forests
- graphs with elements of the same kind (nodes or edges) that have different types (for example: bipartite/k-partite graphs, multimodal graphs)
- hypergraphs
common.graph
does not allow graphs with both directed and undirected edges.
The Graphs class provides some basic utilities (for example, copying and comparing graphs).
Graph Types
There are three top-level graph interfaces, that are distinguished by their representation of edges: Graph
, ValueGraph
, and Network
. These are sibling types, i.e., none is a subtype of any of the others.
Each of these "top-level" interfaces extends SuccessorsFunction andPredecessorsFunction. These interfaces are meant to be used as the type of a parameter to graph algorithms (such as breadth first traversal) that only need a way of accessing the successors/predecessors of a node in a graph. This is especially useful in cases where the owner of a graph already has a representation that works for them and doesn't particularly want to serialize their representation into a common.graph
type just to run one graph algorithm.
Graph
Graph is the simplest and most fundamental graph type. It defines the low-level operators for dealing with node-to-node relationships, such assuccessors(node)
, adjacentNodes(node)
, and inDegree(node)
. Its nodes are first-class unique objects; you can think of them as analogous to Map
keys into the Graph
internal data structures.
The edges of a Graph
are completely anonymous; they are defined only in terms of their endpoints.
Example use case: Graph<Airport>
, whose edges connect the airports between which one can take a direct flight.
ValueGraph
ValueGraph has all the node-related methods that Graph does, but adds a couple of methods that retrieve a value for a specified edge.
The edges of a ValueGraph
each have an associated user-specified value. These values need not be unique (as nodes are); the relationship between aValueGraph
and a Graph
is analogous to that between a Map
and a Set
; aGraph
's edges are a set of pairs of endpoints, and a ValueGraph
's edges are a map from pairs of endpoints to values.)
ValueGraph provides an asGraph()
method which returns a Graph
view of the ValueGraph
. This allows methods which operate on Graph
instances to function for ValueGraph
instances as well.
Example use case: ValueGraph<Airport, Distance>
, whose edge values represent the distance between the two Airport
s that the edge connects.
Network
Network has all the node-related methods that Graph
does, but adds methods that work with edges and node-to-edge relationships, such as outEdges(node)
,incidentNodes(edge)
, and edgesConnecting(nodeU, nodeV)
.
The edges of a Network
are first-class (unique) objects, just as nodes are in all graph types. The uniqueness constraint for edges allows Network to natively support parallel edges, as well as the methods relating to edges and node-to-edge relationships.
Network provides an asGraph()
method which returns a Graph
view of theNetwork
. This allows methods which operate on Graph
instances to function for Network
instances as well.
Example use case: Network<Airport, Flight>
, in which the edges represent the specific flights that one can take to get from one airport to another.
Choosing the right graph type
The essential distinction between the three graph types is in their representation of edges.
Graph has edges which are anonymous connections between nodes, with no identity or properties of their own. You should use Graph
if each pair of nodes is connected by at most one edge, and you don't need to associate any information with edges.
ValueGraph has edges which have values (e.g., edge weights or labels) that may or may not be unique. You should use ValueGraph
if each pair of nodes is connected by at most one edge, and you need to associate information with edges that may be the same for different edges (for example, edge weights).
Network has edges which are first-class unique objects, just as nodes are. You should use Network
if your edge objects are unique, and you want to be able to issue queries that reference them. (Note that this uniqueness allowsNetwork
to support parallel edges.)
Building graph instances
The implementation classes that common.graph
provides are not public, by design. This reduces the number of public types that users need to know about, and makes it easier to navigate the various capabilities that the built-implementations provide, without overwhelming users that just want to create a graph.
To create an instance of one of the built-in implementations of a graph type, use the corresponding Builder
class: GraphBuilder, ValueGraphBuilder, or NetworkBuilder. Examples:
// Creating mutable graphs MutableGraph graph = GraphBuilder.undirected().build();
MutableValueGraph<City, Distance> roads = ValueGraphBuilder.directed() .incidentEdgeOrder(ElementOrder.stable()) .build();
MutableNetwork<Webpage, Link> webSnapshot = NetworkBuilder.directed() .allowsParallelEdges(true) .nodeOrder(ElementOrder.natural()) .expectedNodeCount(100000) .expectedEdgeCount(1000000) .build();
// Creating an immutable graph ImmutableGraph countryAdjacencyGraph = GraphBuilder.undirected() .immutable() .putEdge(FRANCE, GERMANY) .putEdge(FRANCE, BELGIUM) .putEdge(GERMANY, BELGIUM) .addNode(ICELAND) .build();
- You can get an instance of a graph
Builder
in one of two ways:- calling the static methods
directed()
orundirected()
. Each Graph instance that theBuilder
provides will be directed or undirected. - calling the static method
from()
, which gives you aBuilder
based on an existing graph instance.
- calling the static methods
- After you've created your
Builder
instance, you can optionally specify other characteristics and capabilities. - Building mutable graphs
- You can call
build()
on the sameBuilder
instance multiple times to build multiple graph instances with the same configuration. - You don't need to specify the element type(s) on the
Builder
; specifying them on the graph type itself is sufficient. - The
build()
method returns aMutable
subtype of the associated graph type, which provides mutation methods; more on this in"Mutable and Immutable graphs", below.
- You can call
- Building immutable graphs
- You can call
immutable()
on the sameBuilder
instance multiple times to create multipleImmutableGraph.Builder
instances with the same configuration. - You need to specify the element type(s) on the
immutable
call.
- You can call
Builder constraints vs. optimization hints
The Builder
types generally provide two types of options: constraints and optimization hints.
Constraints specify behaviors and properties that graphs created by a givenBuilder
instance must satisfy, such as:
- whether the graph is directed
- whether this graph allows self-loops
- whether this graph's edges are sorted
and so forth.
Optimization hints may optionally be used by the implementation class to increase efficiency, for example, to determine the type or initial size of internal data structures. They are not guaranteed to have any effect.
Each graph type provides accessors corresponding to its Builder
-specified constraints, but does not provide accessors for optimization hints.
Mutable
and Immutable
graphs
Mutable*
types
Each graph type has a corresponding Mutable*
subtype: MutableGraph,MutableValueGraph, and MutableNetwork. These subtypes define the mutation methods:
- methods for adding and removing nodes:
addNode(node)
andremoveNode(node)
- methods for adding and removing edges:
- MutableGraph
*putEdge(nodeU, nodeV)
*removeEdge(nodeU, nodeV)
- MutableValueGraph
*putEdgeValue(nodeU, nodeV, value)
*removeEdge(nodeU, nodeV)
- MutableNetwork
*addEdge(nodeU, nodeV, edge)
*removeEdge(edge)
- MutableGraph
This is a departure from the way that the Java Collections Framework--and Guava's new collection types--have historically worked; each of those types includes signatures for (optional) mutation methods. We chose to break out the mutable methods into subtypes in part to encourage defensive programming: generally speaking, if your code only examines or traverses a graph and does not mutate it, its input should be specified as on Graph, ValueGraph, orNetwork rather than their mutable subtypes. On the other hand, if your code does need to mutate an object, it's helpful for your code to have to call attention to that fact by working with a type that labels itself "Mutable".
Since Graph, etc. are interfaces, even though they don't include mutation methods, providing an instance of this interface to a caller does not guarantee that it will not be mutated by the caller, as (if it is in fact aMutable*
subtype) the caller could cast it to that subtype. If you want to provide a contractual guarantee that a graph which is a method parameter or return value cannot be modified, you should use the Immutable
implementations; more on this below.
Immutable*
implementations
Each graph type also has a corresponding Immutable
implementation. These classes are analogous to Guava's ImmutableSet
, ImmutableList
,ImmutableMap
, etc.: once constructed, they cannot be modified, and they use efficient immutable data structures internally.
Unlike the other Guava Immutable
types, however, these implementations do not have any method signatures for mutation methods, so they don't need to throwUnsupportedOperationException
for attempted mutates.
You create an instance of an ImmutableGraph
, etc. in one of two ways:
Using GraphBuilder
:
ImmutableGraph immutableGraph1 = GraphBuilder.undirected() .immutable() .putEdge(FRANCE, GERMANY) .putEdge(FRANCE, BELGIUM) .putEdge(GERMANY, BELGIUM) .addNode(ICELAND) .build();
Using ImmutableGraph.copyOf()
:
ImmutableGraph<Integer> immutableGraph2 = ImmutableGraph.copyOf(otherGraph);
Immutable graphs are always guaranteed to provide a stable incident edge order. If the graph is populated using GraphBuilder
, then the incident edge order will be insertion order where possible (see ElementOrder.stable() for more info). When using copyOf
, then the incident edge order will be the order in which they are visited during the copy process.
Guarantees
Each Immutable*
type makes the following guarantees:
- shallow immutability: elements can never be added, removed or replaced (these classes do not implement the
Mutable*
interfaces) - deterministic iteration: the iteration orders are always the same as those of the input graph
- thread safety: it is safe to access this graph concurrently from multiple threads
- integrity: this type cannot be subclassed outside this package (which would allow these guarantees to be violated)
Treat these classes as "interfaces", not implementations
Each of the Immutable*
classes is a type offering meaningful behavioral guarantees -- not merely a specific implementation. You should treat them as interfaces in every important sense of the word.
Fields and method return values that store an Immutable*
instance (such asImmutableGraph
) should be declared to be of the Immutable*
type rather than the corresponding interface type (such as Graph
). This communicates to callers all of the semantic guarantees listed above, which is almost always very useful information.
On the other hand, a parameter type of ImmutableGraph
is generally a nuisance to callers. Instead, accept Graph
.
Warning: as noted elsewhere, it is almost always a bad idea to modify an element (in a way that affects its equals()
behavior) while it is contained in a collection. Undefined behavior and bugs will result. It's best to avoid using mutable objects as elements of anImmutable*
instance at all, as users may expect your "immutable" object to be deeply immutable.
Graph elements (nodes and edges)
Elements must be useable as Map
keys
The graph elements provided by the user should be thought of as keys into the internal data structures maintained by the graph implementations. Thus, the classes used to represent graph elements must have equals()
and hashCode()
implementations that have, or induce, the properties listed below.
Uniqueness
If A
and B
satisfy A.equals(B) == true
then at most one of the two may be an element of the graph.
Consistency between hashCode()
and equals()
hashCode()
must be consistent with equals()
as defined byObject.hashCode().
Ordering consistency with equals()
If the nodes are sorted (for example, via GraphBuilder.orderNodes()
), the ordering must be consistent with equals()
, as defined by Comparator andComparable.
Non-recursiveness
hashCode
and equals()
must not recursively reference other elements, as in this example:
// DON'T use a class like this as a graph element (or Map key/Set element) public final class Node { T value; Set<Node> successors;
public boolean equals(Object o) { Node other = (Node) o; return Objects.equals(value, other.value) && Objects.equals(successors, other.successors); }
public int hashCode() { return Objects.hash(value, successors); } }
Using such a class as a common.graph
element type (e.g., Graph<Node<T>>
) has these problems:
- redundancy: the implementations of
Graph
provided by thecommon.graph
library already store these relationships - inefficiency: adding/accessing such elements calls
equals()
(and possiblyhashCode()
), which require O(n) time - infeasibility: if there are cycles in the graph,
equals()
andhashCode()
may never terminate
Instead, just use the T
value itself as the node type (assuming that theT
values are themselves valid Map
keys).
Elements and mutable state
If graph elements have mutable state:
- the mutable state must not be reflected in the
equals()/hashCode()
methods (this is discussed in theMap
documentation in detail) - don't construct multiple elements that are equal to each other and expect them to be interchangeable. In particular, when adding such elements to a graph, you should create them once and store the reference if you will need to refer to those elements more than once during creation (rather than passing
new MyMutableNode(id)
to eachadd*()
call).
If you need to store mutable per-element state, one option is to use immutable elements and store the mutable state in a separate data structure (e.g. an element-to-state map).
Elements must be non-null
The methods that add elements to graphs are contractually required to reject null elements.
Graph traversal
The Traverser class provides methods for breadth-first and depth-first traversals of simple graph-like structures, including but not limited to thecommon.graph
's graph types.
Let's say you want to look for a file in a file system. It may be impractical to put all file paths into an in-memory graph type. However, Traverser works with a functional interface (SuccessorsFunction, also implemented by the graph types):
Iterable
stream(iterable).filter(path -> containsText(path, query)).findFirst();
The example above is (roughly) how Files.fileTraverser() andMoreFiles.fileTraverser() are implemented for the local filesystem, returning Traverser<java.io.File>
and Traverser<java.nio.file.Path>
respectively.
Utility functions
Utility functions live in Graphs.java. Some examples:
- Graphs.transpose() reverses the direction of every edge:
ValueGraph<Country, Money> importGraph = Graphs.transpose(exportGraph); - Graphs.transitiveClosure() returns a graph with an edge connecting node
A
to nodeB
if nodeB
is reachable from nodeA
:
Graph reachableCountriesGraph = Graphs.transitiveClosure(countryAdjacencyGraph);
reachableCountriesGraph.hasEdgeConnecting(SPAIN, CHINA); // = true
Library contracts and behaviors
This section discusses behaviors of the built-in implementations of thecommon.graph
types.
Mutation
You can add an edge whose incident nodes have not previously been added to the graph. If they're not already present, they're silently added to the graph:
MutableGraph graph = GraphBuilder.directed().build(); // graph is empty graph.putEdge(1, 2); // this adds 1 and 2 as nodes of this graph, and puts // an edge between them if (graph.nodes().contains(1)) { // evaluates to "true" ... }
Graph equals()
and graph equivalence
As of Guava 22, common.graph
's graph types each define equals()
in a way that makes sense for the particular type:
Graph.equals()
defines twoGraph
s to be equal if they have the same node and edge sets (that is, each edge has the same endpoints and same direction in both graphs).ValueGraph.equals()
defines twoValueGraph
s to be equal if they have the same node and edge sets, and equal edges have equal values.Network.equals()
defines twoNetwork
s to be equal if they have the same node and edge sets, and each edge object connects the same nodes in the same direction (if any).
In addition, for each graph type, two graphs can be equal only if their edges have the same directedness (both graphs are directed or both are undirected).
Of course, hashCode()
is defined consistently with equals()
for each graph type.
If you want to compare two Network
s or two ValueGraph
s based only on connectivity, or to compare a Network
or a ValueGraph
to a Graph
, you can use the Graph
view that Network
and ValueGraph
provide:
Graph graph1, graph2; ValueGraph<Integer, Double> valueGraph1, valueGraph2; Network<Integer, MyEdge> network1, network2;
// compare based on nodes and node relationships only if (graph1.equals(graph2)) { ... } if (valueGraph1.asGraph().equals(valueGraph2.asGraph())) { ... } if (network1.asGraph().equals(graph1.asGraph())) { ... }
// compare based on nodes, node relationships, and edge values if (valueGraph1.equals(valueGraph2)) { ... }
// compare based on nodes, node relationships, and edge identities if (network1.equals(network2)) { ... }
Accessor methods
Accessors which return collections
- will throw
IllegalArgumentException
if passed an element that is not in the graph - return unmodifiable views of the graph that will update as the graph updates
- iterating over view elements while modifying the graph is not supported and may result in throwing
ConcurrentModificationException
- iterating over view elements while modifying the graph is not supported and may result in throwing
- will return empty collections if their inputs are valid but no elements satisfy the request (for example:
graph.adjacentNodes(node)
will return an empty collection ifnode
is ingraph
but has no adjacent nodes ingraph
).
Views returned by accessors whose arguments are then removed from the graph will throw IllegalStateException
if they are accessed in any way, with the following exceptions:
view.equals(view)
(self-equals) will return trueview.hashCode()
is unchanged (does not throw)- if the argument is then re-added to the graph, the behavior of the view after that point is undefined
Example:
MutableGraph graph; ... Set view = graph.adjacentNodes(node); // accessor based on 'node' Set copy = ImmutableSet.copyOf(view); graph.removeNode(node); // invalidates 'view' view.size(); // throws IllegalStateException view.equals(view); // evaluates to 'true' view.equals(copy); // throws IllegalStateException copy.equals(view); // throws IllegalStateException view.hashCode(); // does not throw
graph.addNode(node); // the behavior of 'view' is now undefined
While some Java Collection Framework methods such as contains()
take Object
parameters rather than the appropriate generic type specifier, as of Guava 22, the common.graph
methods take the generic type specifiers to improve type safety.
Synchronization
It is up to each graph implementation to determine its own synchronization policy. By default, undefined behavior may result from the invocation of any method on a graph that is being mutated by another thread.
Generally speaking, the built-in mutable implementations provide no synchronization guarantees, but the Immutable*
classes (by virtue of being immutable) are thread-safe.
Element objects
The node, edge, and value objects that you add to your graphs are irrelevant to the built-in implementations; they're just used as keys to internal data structures. This means that nodes/edges may be shared among graph instances.
By default, node and edge objects are insertion-ordered (that is, are visited by the Iterator
s for nodes()
and edges()
in the order in which they were added to the graph, as with LinkedHashSet
).
Notes for implementors
Storage models
common.graph
supports multiple mechanisms for storing the topology of a graph, including:
- the graph implementation stores the topology (for example, by storing a
Map<N, Set<N>>
that maps nodes onto their adjacent nodes); this implies that the nodes are just keys, and can be shared among graphs - the nodes store the topology (for example, by storing a
List<E>
of adjacent nodes); this (usually) implies that nodes are graph-specific - a separate data repository (for example, a database) stores the topology
Note: Multimap
s are not sufficient internal data structures for Graph implementations that support isolated nodes (nodes that have no incident edges), due to their restriction that a key either maps to at least one value, or is not present in the Multimap
.
Accessor behavior
For Guava 33.1.0+, all implementations of the graph types are contractually required to return unmodifiable views that become invalid if their arguments are removed from the graph. See the "Accessor methods" section above for details.
Abstract*
classes
Each graph type has a corresponding Abstract
class: AbstractGraph
, etc.
Implementors of the graph interfaces should, if possible, extend the appropriate abstract class rather than implementing the interface directly. The abstract classes provide implementations of several key methods that can be tricky to do correctly, or for which it's helpful to have consistent implementations, such as:
*degree()
toString()
Graph.edges()
Network.asGraph()
Code examples
Is node
in the graph?
graph.nodes().contains(node);
Is there an edge between nodes u
and v
(that are known to be in the graph)?
In the case where the graph is undirected, the ordering of the arguments u
andv
in the examples below is irrelevant.
// This is the preferred syntax since 23.0 for all graph types. graph.hasEdgeConnecting(u, v);
// These are equivalent (to each other and to the above expression). graph.successors(u).contains(v); graph.predecessors(v).contains(u);
// This is equivalent to the expressions above if the graph is undirected. graph.adjacentNodes(u).contains(v);
// This works only for Networks. !network.edgesConnecting(u, v).isEmpty();
// These work only if "network" has at most a single edge connecting u to v. network.edgeConnecting(u, v).isPresent(); // Java 8+ only network.edgeConnectingOrNull(u, v) != null;
// These work only for ValueGraphs. valueGraph.edgeValue(u, v).isPresent(); // Java 8+ only valueGraph.edgeValueOrDefault(u, v, null) != null;
Basic Graph
example
ImmutableGraph graph = GraphBuilder.directed() .immutable() .addNode(1) .putEdge(2, 3) // also adds nodes 2 and 3 if not already present .putEdge(2, 3) // no effect; Graph does not support parallel edges .build();
Set successorsOfTwo = graph.successors(2); // returns {3}
Basic ValueGraph example
MutableValueGraph<Integer, Double> weightedGraph = ValueGraphBuilder.directed().build(); weightedGraph.addNode(1); weightedGraph.putEdgeValue(2, 3, 1.5); // also adds nodes 2 and 3 if not already present weightedGraph.putEdgeValue(3, 5, 1.5); // edge values (like Map values) need not be unique ... weightedGraph.putEdgeValue(2, 3, 2.0); // updates the value for (2,3) to 2.0
Basic Network example
MutableNetwork<Integer, String> network = NetworkBuilder.directed().build(); network.addNode(1); network.addEdge("2->3", 2, 3); // also adds nodes 2 and 3 if not already present
Set successorsOfTwo = network.successors(2); // returns {3} Set outEdgesOfTwo = network.outEdges(2); // returns {"2->3"}
network.addEdge("2->3 too", 2, 3); // throws; Network disallows parallel edges // by default network.addEdge("2->3", 2, 3); // no effect; this edge is already present // and connecting these nodes in this order
Set inEdgesOfFour = network.inEdges(4); // throws; node not in graph
Traversing an undirected graph node-wise
// Return all nodes reachable by traversing 2 edges starting from "node" // (ignoring edge direction and edge weights, if any, and not including "node"). Set getTwoHopNeighbors(Graph graph, N node) { Set twoHopNeighbors = new HashSet<>(); for (N neighbor : graph.adjacentNodes(node)) { twoHopNeighbors.addAll(graph.adjacentNodes(neighbor)); } twoHopNeighbors.remove(node); return twoHopNeighbors; }
Traversing a directed graph edge-wise
// Update the shortest-path weighted distances of the successors to "node" // in a directed Network (inner loop of Dijkstra's algorithm) // given a known distance for {@code node} stored in a {@code Map<N, Double>}, // and a {@code Function<E, Double>} for retrieving a weight for an edge. void updateDistancesFrom(Network<N, E> network, N node) { double nodeDistance = distances.get(node); for (E outEdge : network.outEdges(node)) { N target = network.incidentNodes(outEdge).target(); double targetDistance = nodeDistance + edgeWeights.apply(outEdge); if (targetDistance < distances.getOrDefault(target, Double.MAX_VALUE)) { distances.put(target, targetDistance); } } }
FAQ
Why did Guava introduce common.graph
?
The same arguments apply to graphs as to many other things that Guava does:
- code reuse/interoperability/unification of paradigms: lots of things relate to graph processing
- efficiency: how much code is using inefficient graph representations? too much space (e.g. matrix representations)?
- correctness: how much code is doing graph analysis wrong?
- promotion of use of graph as ADT: how many people would be working with graphs if it were easy?
- simplicity: code which deals with graphs is easier to understand if it’s explicitly using that metaphor.
What kinds of graphs does common.graph
support?
This is answered in the "Capabilities" section, above.
common.graph
doesn’t have feature/algorithm X, can you add it?
Maybe. You can email us at guava-discuss@googlegroups.com
or open an issue on GitHub.
Our philosophy is that something should only be part of Guava if (a) it fits in with Guava’s core mission and (b) there is good reason to expect that it will be reasonably widely used.
common.graph
will probably never have capabilities like visualization and I/O; those would be projects unto themselves and don't fit well with Guava's mission.
Capabilities like traversal, filtering, or transformation are better fits, and thus more likely to be included, although ultimately we expect that other graph libraries will provide most capabilities.
Does it support very large graphs (i.e., MapReduce scale)?
Not at this time. Graphs in the low millions of nodes should be workable, but you should think of this library as analogous to the Java Collections Framework types (Map
, List
, Set
, and so on).
How can I define the order of successors(node)
?
Setting incidentEdgeOrder()
to ElementOrder.stable() in the graph builder makes sure that successors(node)
returns the successors of node
in the order that the edges were inserted. This is also true for most other methods that relate to the incident edges of a node, such as (such as incidentEdges(node)
).
Why should I use it instead of something else?
tl;dr: you should use whatever works for you, but please let us know what you need if this library doesn't support it!
The main competitors to this library (for Java) are:JUNG and JGraphT.
JUNG
was co-created by Joshua O'Madadhain (the common.graph
lead) in 2003, and he still maintains it. JUNG is fairly mature and full-featured and is widely used, but has a lot of cruft and inefficiencies. Now that common.graph
has been released externally, he is working on a new version of JUNG
which usescommon.graph
for its data model.
JGraphT
is another third-party Java graph library that’s been around for a while. We're not as familiar with it, so we can’t comment on it in detail, but it has at least some things in common with JUNG
. This library also includes a number ofadapter classesto adapt common.graph
graphs into JGraphT
graphs.
Rolling your own solution is sometimes the right answer if you have very specific requirements. But just as you wouldn't normally implement your own hash table in Java (instead of using HashMap
or ImmutableMap
), you should consider using common.graph
(or, if necessary, another existing graph library) for all the reasons listed above.
Major Contributors
common.graph
has been a team effort, and we've had help from a number of people both inside and outside Google, but these are the people that have had the greatest impact.
- Omar Darwish did a lot of the early implementations, and set the standard for the test coverage.
- James Sexton has been the single most prolific contributor to the project and has had a significant influence on its direction and its designs. He's responsible for some of the key features, and for the efficiency of the implementations that we provide.
- Joshua O'Madadhain started the
common.graph
project after reflecting on the strengths and weaknesses ofJUNG, which he also helped to create. He leads the project, and has reviewed or written virtually every aspect of the design and the code. - Jens Nyman contributed many of the more recent additions such as Traverser and immutable graph builders. He also has a major influence on the future direction of the project.