xpath2.js - Pure JavaScript implementation of XPath 2 query language

About

xpath2.js is a DOM-agnostic open-source XPath 2.0 implementation in JavaScript. Execution engine operates using XML Schema 1.1 data types as prescribed by specification.

Features

• Full XPath 2.0 language support
• Arbitrary tree structure querying with XPath 2.0 language via custom DOMAdapter
• Custom collation support (using StaticContext)
• Custom function support (using StaticContext)
• Variable injection (using DynamicContext)

Installation

npm install xpath2.js

Usage

The simple API implementation lib/index.js provided for reference. Its primary purpose is to demonstrate implementation classes wiring and a simple usable solution.

Basic scenarious with evaluate function

xpath.evaluate(expression, evaluationContext, staticContext, initialScope, DOMAdapter)

Parameters list

Query without a context

const xpath = require("xpath2.js");
const result = xpath.evaluate("1 to 5");
console.log(result); // prints [ 1, 2, 3, 4, 5 ]

Query a document not specifying namespaces

const xpath = require("xpath2.js");
const xmldom = require("xmldom"); // You are free to use any DOM implementation
const document = new xmldom.DOMParser().parseFromString('<test>content</test>');

const result = xpath.evaluate("fn:string(/test/text())", document);
console.log(result); // prints [ 'content' ]

Query a document with namespace resolver

Evaluating expressions over documents that specify namespaces requires namespace resolver to be provided with the query. Take a note that namespace resolver is there to resolve prefixes found in XPath expressions, thus making use of prefixes in expressions scoped to the query, and not to the document.

A namespace resolver is a function that takes single argument String prefix and returns a namespace uri for it.

Exception XPST0081 will be thrown, should any of the prefixes used in expression are left unresolved.

const xpath = require("xpath2.js");
const xmldom = require("xmldom");
const document = new xmldom.DOMParser().parseFromString('<foo><a:bar xmlns:a="http://a">content</a:bar></foo>');
const namespaceResolver = function(prefix) {
    if (prefix == "b")
        return "http://a";
    return null;
};

const result = xpath.evaluate("fn:string(//b:bar/text())", document, namespaceResolver);
console.log(result); // prints [ 'content' ]

Passing a JavaScript variable to the evaluation context

const xpath = require("xpath2.js");

const result = xpath.evaluate("$a + 0.2", null, null, {a: 0.1});
console.log(result); // prints [ 0.3 ]

More challenging scenarious

Using execute function and managing contexts

const xpath = require("xpath2.js");
const xmldom = require("xmldom");
const document = new xmldom.DOMParser().parseFromString('<foo><a:bar xmlns:a="http://a">content</a:bar></foo>');
const namespaceResolver = function(prefix) {
    if (prefix == "b")
        return "http://a";
    return null;
};
const staticContext = xpath.createStaticContext(namespaceResolver);
// Set default function namespace to the one of XPath functions, so "fn" prefix can be dropped in queries
staticContext.defaultFunctionNamespace = "http://www.w3.org/2005/xpath-functions";
const dynamicContext = xpath.createDynamicContext(staticContext, document);
const expression = xpath.compile("string(//b:bar/text())", staticContext);

const result = xpath.execute(expression, dynamicContext);
console.log(result); // prints [ 'content' ]

Note! Dynamic context carries date/time obtained during its creation