XPath snippets

This article provides some XPath code snippets—simple examples of how to a few simple utility functions based on standard interfaces from the DOM Level 3 XPath specification that expose XPath functionality to JavaScript code. The snippets are functions you can use in the real world in your own code.

Node-specific evaluator function

The following custom utility function can be used to evaluate XPath expressions on given XML nodes. The first argument is a DOM node or Document object, while the second is a string defining an XPath expression.

Example: Defining a custom node-specific evaluateXPath() utility function
// Evaluate an XPath expression aExpression against a given DOM node
// or Document object (aNode), returning the results as an array
// thanks wanderingstan at morethanwarm dot mail dot com for the
// initial work.
function evaluateXPath(aNode, aExpr) {
  var xpe = new XPathEvaluator();
  var nsResolver = xpe.createNSResolver(aNode.ownerDocument == null ?
    aNode.documentElement : aNode.ownerDocument.documentElement);
  var result = xpe.evaluate(aExpr, aNode, nsResolver, 0, null);
  var found = [];
  var res;
  while (res = result.iterateNext())
    found.push(res);
  return found;
}

This function uses the new XPathEvaluator() constructor, which is supported in Firefox, Chrome, Opera and Safari, but not in Edge or Internet Explorer. Scripts in a Web document which might be accessed by Edge or Internet Explorer users should replace the call to new XPathEvaluator() with the following fragment:

  // XPathEvaluator is implemented on objects that implement Document
  var xpe = aNode.ownerDocument || aNode;

In that case the creation of the XPathNSResolver can be simplified as:

  var nsResolver = xpe.createNSResolver(xpe.documentElement);

Note however that createNSResolver should only be used if you are sure the namespace prefixes in the XPath expression match those in the document you want to query (and that no default namespace is being used (though see document.createNSResolver for a workaround)). Otherwise, you have to provide your own implementation of XPathNSResolver.

If you are using XMLHttpRequest to read a local or remote XML file into a DOM tree (as described in Parsing and serializing XML), the first argument to evaluateXPath() should be req.responseXML.

Sample usage

Assume we have the following XML document (see also How to Create a DOM tree and Parsing and serializing XML):

Example: An XML document to use with the custom evaluateXPath() utility function
<?xml version="1.0"?>
<people>
  <person first-name="eric" middle-initial="H" last-name="jung">
    <address street="321 south st" city="denver" state="co" country="usa"/>
    <address street="123 main st" city="arlington" state="ma" country="usa"/>
  </person>

  <person first-name="jed" last-name="brown">
    <address street="321 north st" city="atlanta" state="ga" country="usa"/>
    <address street="123 west st" city="seattle" state="wa" country="usa"/>
    <address street="321 south avenue" city="denver" state="co" country="usa"/>
  </person>
</people>

You can now "query" the document with XPath expressions. Although walking the DOM tree can achieve similar results, using XPath expressions is much quicker and more powerful. If you can rely on id attributes, document.getElementById() is still powerful, but it's not nearly as powerful as XPath. Here are some examples.

Example: JavaScript code with the custom evaluateXPath() utility function
// display the last names of all people in the doc
var results = evaluateXPath(people, "//person/@last-name");
for (var i in results)
  alert("Person #" + i + " has the last name " + results[i].value);

// get the 2nd person node
results = evaluateXPath(people, "/people/person[2]");

// get all the person nodes that have addresses in denver
results = evaluateXPath(people, "//person[address/@city='denver']");

// get all the addresses that have "south" in the street name
results = evaluateXPath(people,  "//address[contains(@street, 'south')]");
alert(results.length);

docEvaluateArray

The following is a simple utility function to get (ordered) XPath results into an array, regardless of whether there is a special need for namespace resolvers, etc. It avoids the more complex syntax of document.evaluate() for cases when it is not required as well as the need to use the special iterators on XPathResult (by returning an array instead).

Example: Defining a simple docEvaluateArray​() utility function
// Example usage:
// var els = docEvaluateArray('//a');
// alert(els[0].nodeName); // gives 'A' in HTML document with at least one link

function docEvaluateArray (expr, doc, context, resolver) {
    var i, result, a = [];
    doc = doc || (context ? context.ownerDocument : document);
    resolver = resolver || null;
    context = context || doc;
	
    result = doc.evaluate(expr, context, resolver, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null);
    for(i = 0; i < result.snapshotLength; i++) {
        a[i] = result.snapshotItem(i);
    }
    return a;
}

getXPathForElement

The following function allows one to pass an element and an XML document to find a unique string XPath expression leading back to that element.

Example: Defining a getXPathForElement​​() utility function
function getXPathForElement(el, xml) {
	var xpath = '';
	var pos, tempitem2;
	
	while(el !== xml.documentElement) {		
		pos = 0;
		tempitem2 = el;
		while(tempitem2) {
			if (tempitem2.nodeType === 1 && tempitem2.nodeName === el.nodeName) { // If it is ELEMENT_NODE of the same name
				pos += 1;
			}
			tempitem2 = tempitem2.previousSibling;
		}
		
		xpath = "*[name()='"+el.nodeName+"' and namespace-uri()='"+(el.namespaceURI===null?'':el.namespaceURI)+"']["+pos+']'+'/'+xpath;

		el = el.parentNode;
	}
	xpath = '/*'+"[name()='"+xml.documentElement.nodeName+"' and namespace-uri()='"+(el.namespaceURI===null?'':el.namespaceURI)+"']"+'/'+xpath;
	xpath = xpath.replace(/\/$/, '');
	return xpath;
}

Resources

See also