mavis said:
Could you instruct how to refer to another element if it is not key or
unique by using xpath?
An XPath is a description of how to find information within a document
-- a simple query operation, if you want to look at it that way.
It's up to the environment to say which document; there's no reason your
system can't use an XPath within a doc to refer to another part of that
doc. XSLT stylesheets sometimes do exactly this, examining themselves to
(for example) do a table lookup. In XSLT, any XPath starting with a call
to the no-arguments version of the document() function is a request to
search the stylesheet document.
Of course, actually following the XPath to the node it refers to will
require running an XPath interpreter, giving it the approrpiate context
information. The TrAX APIs are one semi-standardized way of doing that
in Java; DOM Level 3 also introduces its own XPath API.