rdf-path v0.1.1

rdf-path

This package provides classes and utils to describe paths in RDF graphs based on the RDF/JS Data Model and RDF/JS Dataset.

Interfaces

Node

A Node represents a term inside a dataset in a specific named graph. The constructor accepts dataset, graph and term as named parameters which will be assigned to the properties with the same name. The parameters must be provided just like in the following description.

Properties:

• dataset: The dataset that contains the term as a RDF/JS DatasetCore object.
• graph: The named graph that contains the term as a RDF/JS NamedNode object.
• term: The term as a RDF/JS Term object.

Methods:

• equals(other): Compares this Node with other using equals for each property.

Edge

An Edge represents a connection from one Node to another with a specific predicate in a specific direction. A dataset, dir and quad is used to store that information. The same parameters are required by the constructor. The parameters must be provided just like in the following description.

Properties:

• dataset: The dataset that contains the quad as a RDF/JS DatasetCore object.
• dir: The direction of the connection which can be out for subject -> object or in for object -> subject.
• quad: The quad that contains the connection information as RDF/JS Quad.
• start: The start Node of the edge.
• end: The end Node of the edge.

Methods:

• equals(other): Compares this Edge with other using string compare for dir property and equals for the dataset and quad properties.

Path

A Path represents a chain of Edges connected to each other. A dataset and an array of edges are used to store that information. The dataset parameter can be given in the constructor. If the parameter is not given, it will be assigned when the first Edge is appended.

Properties:

• dataset: The dataset that contains the edges as a RDF/JS DatasetCore object.
• edges: The edges as an array of Edge.
• length: The number of edges in the path.
• out: True if there is an Edge in the Path with the direction out.
• in: True if there is an Edge in the Path with the direction in.
• start: The first Node of the Path.
• end: The last Node of the Path.

Methods:

• toString(): Returns the terms of the Nodes separated by comma as a string.
• clone(): Returns a new Path object with the same dataset and a copy of the edges array.
• node(index): Returns the Node with the given index, start counting from start.
• append(edge): Appends an Edge to the Path. Returns the Path itself.

Functions

neighbors(node, options)

Searches for all neighbors of the given node and returns them as a Set of Edges from node to the neighbor. The search behavior can be configured using the following options:

• blacklist: A blacklist for the predicates of the edges given as a TermSet of RDF/JS Term.
• dirIn: Boolean value to enable the search in the direction object -> subject. The default value is false (disabled).
• dirOut: Boolean value to enable the search in the direction subject -> object. The default value is true (enabled).
• followLiterals: Boolean value to enable the search for literal objects. The default value is false (disabled).
• whitelist: A whitelist for the predicates of the edges given as a TermSet of RDF/JS Term.

shortest (start, end, options)

Searches for the shortest path between start and end and returns it as Path object. If no path was found null is returned. The search behavior can be configured using the following options:

• blacklist: A blacklist for the predicates of the edges given as a TermSet of RDF/JS Term.
• cutoff: Stop searching for more edges if the path has already cutoff number of edges. The default value is Number.MAX_SAFE_INTEGER.
• dirIn: Boolean value to enable the search in the direction object -> subject. The default value is false (disabled).
• dirOut: Boolean value to enable the search in the direction subject -> object. The default value is true (enabled).
• followLiterals: Boolean value to enable the search for literal objects. The default value is false (disabled).
• whitelist: A whitelist for the predicates of the edges given as a TermSet of RDF/JS Term.

Usage

All classes and functions can be imported directly from the file matching the class or function name in the root directory. The code below shows how Node and shortest are imported.

// import the RDF/JS data model to create NamedNode objects
const rdf = {...require('@rdfjs/data-model'), ...require('@rdfjs/dataset')}

// each class and function of the rdf-path package can be imported separated using the path
const Node = require('rdf-path/Node')
const shortest = require('rdf-path/shortest')

// this will just create the dataset, it must be also populated (not included in this example)
const dataset = rdf.dataset()
 
// create the start and end node
const start = new Node({ dataset, term: rdf.namedNode('http://example.org/start') })
const end = new Node({ dataset, term: rdf.namedNode('http://example.org/end') })

// search for the shortest path between start and end
const path = shortest(start, end)

// write the chain of nodes to the console
console.log(path.toString())

It's also possible to import the classes and functions from the root object exported by the package. For the Node class that would look like this:

const { Node } = require('rdf-path')

Examples

The examples folder contains an example based on the (The Big Bang Theory dataset)https://github.com/zazuko/tbbt-ld. It shows how to search for the shortest path between a named node and a literal.