Extract <svg> element by using document.evaluate()?

Question

I am trying to use document.evaluate() to extract certain child elements out of a <svg> element. But I have problems by just extracting the <svg> itself. I can extract everything up to the <svg>, but no further. For example this works well :

document.evaluate('//*[@id="chart"]/div/div[1]/div', document, null, XPathResult.ANY_UNORDERED_NODE_TYPE, null).singleNodeValue)

This gives me (shortened) :

<div style="position: absolute; left: 0px; top: 0px; width: 100%; height: 100%;" aria-label="Et diagram.">
    <svg width="600" height="400" aria-label="Et diagram." style="overflow: hidden;"></svg>
    <div aria-label="En repræsentation i tabelformat af dataene i diagrammet." style="position: absolute; left: -10000px; top: auto; width: 1px; height: 1px; overflow: hidden;"></div>
</div>

and then I would assume the simple

//*[@id="chart"]/div/div[1]/div/svg

would give me the <svg> - but nothing works. Have tried all respond types :

for (var type=XPathResult.ANY_TYPE; type<XPathResult.FIRST_ORDERED_NODE_TYPE+1; type ++) {
   console.dir(
      document.evaluate('//*[@id="chart"]/div/div[1]/div/svg', document, null, type, null)
   );   
}    

I get either invalidIteratorState, null singleNodeValue or a snapshotLength of 0.

demo -> http://jsfiddle.net/hw79zp7j/

Is there any limitations of evaluate() I havent heard of, or am I just doing it utterly wrong?

For clarification, my ultimate goal is to be able extract for example a google visualization xAxis <text> elements in a single line of code. As it is now (see demo) I have to iterate through the <svg> by using multiple querySelectorAll / getElementsByTagName etc, which is not very readable, maintenance friendly or elegant. It would be nice just to grab the xpath from google developer tools and reuse it in a single line.

Answer 1

SVG elements are in the namespace http://www.w3.org/2000/svg and to select elements in a namespace with XPath you need to bind a prefix to the namespace URI and use that prefix in your path to qualify the element names e.g. svg:svg. So use a resolver as the third argument of evaluate that maps the prefix you can choose (e.g. svg) to the namespace URI mentioned above.

document.evaluate('//*[@id="chart"]/div/div[1]/div/svg:svg', document, 
                                function(prefix) { 
                                    if (prefix === 'svg') 
                                    { 
                                        return 'http://www.w3.org/2000/svg';
                                    }
                                    else
                                    {
                                        return null;
                                    }
                                }, type, null)