Lists Home |
Date Index |
On Tue, 2004-12-07 at 22:16 -0800, Daniela Florescu wrote:
> > 1. Xquery is not XML
> I was always puzzled by this argument that people use in
> favor of XSLT and against XQuery.
> First, please let me remember you that XSLT is not XML
> in it's entirety EITHER. Do you forget that the path
> expressions remain unparsed strings and appear as
> attribute values !?
I *think* attributes are still part of XML?
I believe XSLT... or the stylesheets I've seen,
are well formed XML too.
> For me this dual representation of XSLT (half pre-parsed XML,
> half unparsed strings) doesn't bring any advantage.
For one, I can generate it with xslt,
and I can parse it using XML tools.
I think that's an advantage.
> I love the *semantics* of templates in XSLT. I find it really useful.
> But I would prefer to also have an angle brackets free syntax for it.
I'm sure someone will or has written a round bracket form of XSLT.
> As such (with the current XSLT syntax) it makes it harder for me
> as a developer (I tend to dislike programming with angle brackets
> --- it's not readable enough), and due to the fact it is XML only half
> it doesn't make it easier for automatic processing either.
I guess I'm the opposite. Most people I've come across prefer the rnc
format of relax-ng, I prefer the rng format.
> So Dave, we had this argument a couple of times, and I never understood
> your answer: what exact benefit do you get from the the fact that XSLT
> represented in it's current XML syntax ?
> And why don't you get the same benefits from XQueryX (the pure XML
> variant of XQuery) ?
I am fairly familiar with XML processing. I generate and process
XSLT as XML instances.
I don't get the benefits from XQueryX because,
I've never used it.
Never found a use for it,
and ... I guess,
am very unlikely to use it..
unless I find a use for an XML database, in which case couldn't I use