Re: [xsl] The identity transform and attributes

[Abel Braaksma <abel.online@xxxxxxxxx>]

David Carlisle wrote:
> > What's however still beyond me (and maybe I missed it on this 
> > afternoon's very interesting thread) is why @* uses the combined syntax 
> > of * (element) and @ (attribute),
> >     
>
> No, * doesn't mean element it means "principal node kind" which is
> element on all axes except attribute: (where it's attribute and
> namespace:: (ehere it is namespace) this pricipal node kind stuff is
> introduced (I assume) just to give a ratinale for @* to mean whet it
> means, but the particular formulation has the effect that self:: axis
> could only be used with difficulty on attribute nodes in XSLT 1 and
> still in xslt 2 it isn't as natural as one might like.
>
>   
> > (i.e., node() is supposed to be "any" node imo, but this is 
> > arguably not true as it does not include attribute nodes),
> >     
>
> yes it does include attribute nodes,
>
> attribute::node() and self::node() both select attribute nodes.
> node() on its own is short for child::node() which does not select
> attribute nodes because there are never any attributes in the child
> axis, it is teh child axis that does not include attributes, not the
> node() node test.

That's a great insight. Sometimes knowing how to write something is one 
thing, knowing why it is the way it is (and in this sense, understand 
the default axis) makes it clearer, brighter and, most importantly, 
understandable.

Thanks,
Abel

PS: what worries me is that I've received this kind of lecture before 
and apparently it does not really sink in, even though I feel I have 
quite an understanding about XSLT. Ah well, next year perhaps ;)