> There's still too much white space there. For instance consider this
> <query id="A1">
> <description>The children nodes of the root that are
> <XPATH_RESULT xmlns:svg="http://www.w3.org/2000/svg"
> <!-- This example shows the use of namespaces, comments, processing
> instructions, and language attribute. The example has been adapted from
> Example 4-4 in Elliotte Rusty Harold and W. Scott Means, XML in a
> Nutshell, 2nd Edition, O'Reilly. -->
> <!-- This is a processing instruction to tell search-engines whether and
> how they should index this page. -->
> <!-- A namespace for XML Linking Language with prefix xlink and the
> attribute declaring the English language. -->
> <!-- This is the last comment. -->
> In the XPath data model, the result you're indicating includes 3 comment
> nodes and 4 text nodes containing line breaks. There is no way
> /comment() can return any text nodes.
> White space counts in XML, and it is something that a lot of tools get
> wrong. It's very important that conformance tests get this right.
ok, I got the message! :-) I will work around the whitespace problem.
Probably I will format the results as Saxon does.
Thanks for your comments,