Re: negate predicate

On Tue, 14 Dec 2004 14:37:36 GMT, "David Carlisle" <davidc@xxxxxxxxx>
said:
> 
> 
>   I want to select the text of all descendants of a node which *don't*
>   have an attribute called 'id' (and none of the nodes which are
>   descendants of nodes with id attributes).
> 
> By that I think you mean you want to select all text nodes that don't
> have an ancestor with an id attribute, that would be
> 
> //text()[not(ancestor::*/@id)]
> 
> Note by the way that [@id = '']/doesn't test if te element has an id
> attribute, just if it is empty <foo id=""/> woul dtest true, you want
> [not(@id)] to test for the attribute itself rather than its value.
> 
Thanks for the tips.

I still can't quite get it to work. Perhaps I'd better explain in more
detail. I'm actually working with XQuery (though the problem is more an
XPath more than an XQuery one). My XQuery statements select all the
'hierachical nodes' (i.e. the sections and subsections) and, for each
one, I want to extract all the text of its child nodes (those /without/
and 'id' attribute) but not the text of its child /hierarchical/ nodes
(those /with/ an 'id' attribute).

for $section in collection("...")//section
    $section//text()[not(ancestor::*/@id)]

seems to cause the script to crash :-o

and

for $section in collection("...")//section
    $section//*[not(@id)]/text()

will select all the text of the $section, but also of its child
'subsection' etc. elements (which I don't want).

I'll carry on fiddling.

Cheers,
Richard
-- 
  Richard Lewis
  richardlewis@xxxxxxxxxxxxxx