Re: DOCBOOK-APPS: Re: needing clarification about XSL transformation

["Robert P. J. Day" <rpjday at mindspring dot com>]

On Mon, 3 Mar 2003, Jirka Kosek wrote:

> "Robert P. J. Day" wrote:
> 
> > which agrees with tidwell, but disagrees with kay, and definitely
> > disagrees with http://www.w3.org/TR/xslt, which states, (Section 5.2)
> > 
> >   "node() matches any node other than an attribute node or the
> >    root node"
> 
> It is correct statement. Note that XPath expression node() (not node
> test alone, whole XPath expression) is shorthand of child::node(). You
> can't select attribute this way as you aren't using attribute:: axis.
> You can't also select root node, because it hasn't parent.

ok, now we're making progress.  so

1) technically, comments and processing-instructions will be matched
   by node(), right?  (that is, those two nodes are in fact child nodes)

2) based on the answer to 1), if i have a template matching "node()"
   and the built-in rule for "comment()", which one will be used
   to process comments?  my general "node()" rule?  or the more
   specific built-in "comment()" rule?  IOW, if i want to do something
   with either comments or processing instructions, must i add a
   template that matches *exactly* a comment or procssing-instruction()?

basically, if a comment node matches node(), i'd like to think that
it would be processed by *my* template using "node()", even if there
is a more specific built-in for "comment()".

rday