wabstractgraph v0.3.33

wAbstractGraph

Collection of abstract data structures and algorithms to process graphs. The module does not bound to any specific format of a graph, so providing adapters toy may use it with anyone. It implements depth-first search, breadth-first search, extracting strongly connected components, topological sort, shortest path search, and others.

Concepts

Graph :: set of nodes and set of edges or arcs connecting some or all nodes. Incident edges :: of the node, are edges connected to the node. Connected nodes :: nodes are connected if them have edge connecting both of them. Reachable node :: node v is reachable from u if there is a path from v to u. DFS :: depth-first search. BFS :: breadth-first search. DAG :: directed acycled graph. SCC :: Strongly connected components. Node degree :: total number of incoming and outgoint edges of the node. Indegree :: of the node is number of incoming edges. Outdegree :: of the node is number of outgoing edges. Sink node :: node with zero outdegree. Source node :: node with zero indegree. Universal node :: node connected to all nodes of the graph. Terminal node :: pendant node :: leaf node :: node with degree of one. Neighborhood :: is an enduced subgraph of the graph, formed by all nodes adjacent to v. Neigbour nodes :: nodes which are connected to the node. Low-link value of a node :: smallest node id reachable from the node. Topological ordering :: linear ordered DAG. Topological sort :: algorithm of linear ordering of DAG. Distance between nodes :: minimal number of edges to get from one given node to another given node. Distance layers :: array of arrays of nodes. First layer has origin or zero-distance set of nodes. Second layer has nodes on distance one from origin. And os on.

Sample

require( '..' );
let _ = wTools;

/*
This example shows how to create a simple graph.
Graph :: set of nodes( vertices ) and set of edges or arcs connecting some or all nodes.
*/

/*
Define a graph of arbitrary structure.
Strcuture of nodes is arbitrary. It could even be instance of a primitive type.
Group of nodes should have handlers which should return lists of neighbour nodes.
*/

var a = { name : 'a', nodes : [] } // 1
var b = { name : 'b', nodes : [] } // 2
var c = { name : 'c', nodes : [] } // 3

// declare lists neighbour nodes

a.nodes.push( b ); // add edge between nodes a and b
b.nodes.push( c ); // add edge between nodes b and c

/* declare the graph */

var sys = new _.graph.AbstractGraphSystem(); // declare sysyem of graphs
var group = sys.groupMake(); // declare group of nodes
group.nodesAdd([ a, b, c ]); // add nodes to the group

console.log( group.nodesExportInfo() ); // print information about nodes relation

/*
    1 : 2
    2 : 3
    3 :
*/

Try out

npm install
node sample/Sample.s