My primary interest is to feed a string of XML (or better still HTML) to a Pike library function to generate a sloppy DOM, and then feed the DOM and a string of XPath, say "id('blog')//*[contains(@class,'post')] /ancestor::li[1]" to an XPath evaluator, and get an array or list of all DOM node matches for said XPath expression.
"Nodes can be selected using a subset of XPath." sounded promising, as Roxen CMS was capable of a rather friendly subset of XPath (including, IIRC, mostly everything except document(), and maybe other silliness), and I'm curious of whether it might handle some, or all, of the above.
(I have a feeling I should probably use rhino to do that kind of thing these days, but it would be cool if pike handled it, as I have a few pike hacks that would benefit from it.)