Using Xpath to get name with same name at different depths

[nick_tucker]

Hi,
I am very new to XML and XPATH. I have made a sample XML fileto ask
my question.

<?xml version="1.0" encoding="utf-8"?>
<Test>
<A>
<B>
<C>
<D>
<NAME>Jack</NAME>
</D>
</C>
</B>
</A>
<J>
<NAME>James</NAME>
</J>
<X>
<Y>
<NAME>Tom</NAME>
</Y>
</X>
<Z>
<NAME>Peter
<W>
<NAME>Harry</NAME>
</W>
</NAME>
</Z>
</Test>

From the XML above how would I get a set of all the NAME xml nodes

using XPATH (Note that the last name node contains another name node)??
The true reson I need to do this sort of query is to get a list of
Table nodes (that can contain other Tables) as I have run some html
through SGMLReader.
Also how would I just get the lowest depth name nodes i.e. so peter is
not returned but jack,JAmes, Tom and Harry are.

Thanks for any help,
Nick

[Martin Honnen]

ni*********@hot mail.com wrote:

From the XML above how would I get a set of all the NAME xml nodes using XPATH (Note that the last name node contains another name node)??

XPath (1.0) can only select existing nodes, to select all NAME element
nodes at all levels use
//NAME
Also how would I just get the lowest depth name nodes i.e. so peter is
not returned but jack,JAmes, Tom and Harry are.

The NAME elements are not leaf nodes, they have text child nodes.
If you want to select all NAME element nodes which do not have any
element child nodes you can use e.g.
//NAME[not(*)]
If you really want the leaf text nodes inside of these elements then you
could use e.g.
//NAME[not(*)]/text()

Check out an XPath tutorial, there are lots on the web, XPath 1.0 is
years old.

One example introduction is here:
<http://msdn.microsoft. com/library/default.asp?url =/library/en-us/xmlsdk/html/6da1b6e3-256e-4919-8848-53b425f72ed1.as p>
--

Martin Honnen --- MVP XML
http://JavaScript.FAQTs.com/

[Peter Flynn]

ni*********@hot mail.com wrote:

Hi,
I am very new to XML and XPATH. I have made a sample XML fileto ask
my question.

<?xml version="1.0" encoding="utf-8"?>
<Test>
<A>
<B>
<C>
<D>
<NAME>Jack</NAME>
</D>
</C>
</B>
</A>
<J>
<NAME>James</NAME>
</J>
<X>
<Y>
<NAME>Tom</NAME>
</Y>
</X>
<Z>
<NAME>Peter
Don't do this. Mixing character data content and element content in the
same element is something you only do in normal text documents (eg HTML
paragraphs and the equivalent where you have continuous text, not in XML
data.
<W>
<NAME>Harry</NAME>
</W>
</NAME>
</Z>
</Test>

From the XML above how would I get a set of all the NAME xml nodes using XPATH (Note that the last name node contains another name node)??

Design your document so that names really are all kept in NAME elements
with no mixing.
The true reson I need to do this sort of query is to get a list of
Table nodes (that can contain other Tables) as I have run some html
through SGMLReader.
Also how would I just get the lowest depth name nodes
By how much the lowest? The single lowest one (Jack) or some collection
of the N lowest ones?
i.e. so peter is
not returned but jack,JAmes, Tom and Harry are.

If you redesign it so the last bit goes

<Z>
<NAME>Peter</NAME>
<W>
<NAME>Harry</NAME>
</W>
</Z>

then the XPath expression count(ancestor: :*) will return the depth for
each NAME node.

///Peter
--
XML FAQ: http://xml.silmaril.ie/