RE: [xsl] Re: RE: Postional predicates de-mystified

> On XSL-List, Mike Kay wrote:
> > No. The attribute value is used directly as the separator. It's a
> > string, not an expression. If you want to compute it, use an
> > attribute value template. We decided to optimize for the common case
> > of a constant string.
> 
> On this topic, I think that it might be worth adding a note in the
> definition of xsl:value-of to state what focus is used to evaluate an
> attribute value template in the separator attribute.
> 
> The section on the 'effective value' of attribute value templates does
> not describe this (which is understandable, given that it doesn't
> really apply in "normal" attribute value templates). It is made more
> explicit in other places, for example the AVTs on xsl:sort.

I think that technically, it's covered by the general rule in the last
paragraph (and bullet point) of section 4.2, but it would be helpful to
cross-refer to this from section 4.5 (which describes the general rules for
AVTs). I'm reluctant to describe the rules for each AVT independently, it
gets too repetitive. It's needed on xsl:sort, because as xsl:sort isn't an
instruction, the rules are non-standard.

Mike Kay 

Received on Monday, 7 January 2002 07:51:09 UTC