Re: Restricting the scope of preceding:: and following

On Wed, 2006-01-04 at 10:40, Geert Josten wrote:
> >>Perhaps you are looking something is the line of:
> >>(descendant::*|following-sibling::*|ancestor::*[count(ancestor::s|$sentence) = 
> >>1]/following-sibling::*/descendant::*)/@type
> > 
> > 
> > Not so much that as the fact that it appears to be ignoring the
> > test count(ancestor::s|$sentence)=1 because it is processing
> > all following:: elements in the document instead of being limited
> > to those within the same sentence as the context element.
> 
> Hmmm, yes, that is probably true. If there is no ancestor::s, then the count is still 1. I've always 
> preferred the variant: generate-id(ancestor::s[1]) = generate-id($sentence). That should give better 
> results..

Oh duuuuh. Of course. Thanks.

///Peter