Java: How do I get an xpath for a org.w3c.dom.Node?

Question

I'm using Java 1.6. Given a org.w3c.dom.Node object, how do I get a String XPath expression for where that node falls in an XML document? It is NOT a given that this node or any of its ancestors/descendants have an id attribute.

This has been answered many times for JavaScript on SO. You might get some use out of looking at some of those answers. Like this one: http://stackoverflow.com/questions/2661818/javascript-get-xpath-of-a-node — Wayne, Dec 07 '11 at 15:16

Adrian Mouat · Accepted Answer · 2012-01-05T21:17:08.580

1

I don't believe there is any easy way via DOM. The problem is that there is not a single canonical way to identify a given node.

Having said that, you might find the the function getXPath from diffxml useful: http://diffxml.cvs.sourceforge.net/viewvc/diffxml/diffxml/src/java/org/diffxml/diffxml/fmes/NodeOps.java?view=markup

It will return an XPath of the form /node()[4]/node()[2] etc.

UPDATE: As diffxml is GPL licensed and getXPath takes a bit of extracting, I released it as XPathGen on github: https://github.com/amouat/XPathGen under an Apache licence.

edited Jan 05 '12 at 21:17

answered Dec 07 '11 at 15:04

Adrian Mouat

44,585
16
110
102

There might not be a *canonical* way to represent a node, but you can always do it accurately by position. – Wayne Dec 07 '11 at 15:17
I use code just like this for debugging purposes. Good stuff. – JohnnyK Dec 07 '11 at 15:41

score 0 · Answer 2 · answered Nov 12 '14 at 15:44

You could pass your node and an empty string to a recursive method such as this one:

static void xpm(Node node, String xpath){
    if (node.getParentNode()==null){
        System.out.println(xpath);
        return;
    }

    xpath = "/"+node.getNodeName()+xpath;
    node = node.getParentNode();
    xpm(node, xpath);
}

Java: How do I get an xpath for a org.w3c.dom.Node?

2 Answers2