[jdom-interest] Element versus Document with XPath

[Jason Hunter]

Your XPath probably includes the name of the element against which you 
were trying to execute the XPath.  That works OK if you're starting at a 
document node (because that elt is its first child), but not if you're 
at the element itself.

Say you have:

<a>
   <b/>
</a>

If you start your XPath at the document, then "a/b" returns "b".  If you 
start at "a", you'll get nothing because "a" has no child named "a" 
which is what "a/b" would be asking for.

-jh-

Mark Hale wrote:

> I was wondering if any of you have ever experienced the same problems.
> If I do an :
> 
> 	Element result = (Element) XPath.selectSingleNode(el, "some
> path");
> 
> I get result=null.  However, if I do:
> 
> 	Document doc = new Document ((Element) el.clone(), new
> DocType("xml"));
> 	Element result = (Element) XPath.selectSingleNode(doc, "some
> path");
> 
> I get the right answer for result.
> 
> Why is it that Xpath doesn't work on the element but does on the
> document?
> 
> Thanks - I am stumped and see this behaviour quite often.
> 
> 	Mark
> 
> 
> _______________________________________________
> To control your jdom-interest membership:
> http://www.jdom.org/mailman/options/jdom-interest/youraddr@yourhost.com
>