RE: Variables and constants

[Mike Brown <mbrown@xxxxxxxxxxxxx>]

John Lam wrote:
> it would also seem to me that the presence of 
> <xsl:call-template> is a compromise that allows
> procedural programmers to adjust to XSLT faster.

I wouldn't say that. I have seen named templates abused before, but if used
as if they are used as extension elements would be, I don't think they
violates any side-effect-free, object-oriented principles. It's nice to have
templates that don't require matching a particular kind of node in the
source tree, but that do act on parameters or regurgitate a common result
tree fragment.

> Come to think about it, doesn't <xsl:for-each> also fit the 
> same bill? I can't see why <xsl:apply-templates> + modes
> can't accomplish the same things that <xsl:for-each> can.

Sure they can, but why should they have to? :)


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list