[XSL-LIST Mailing List Archive Home]
RE: [xsl] XSLT 2.1
Subject: RE: [xsl] XSLT 2.1|
From: "Scott Trenda" <Scott.Trenda@xxxxxxxx>
Date: Fri, 31 Oct 2008 15:12:25 -0500
This is a fantastic analogy. Kudos, Wendell.
From: Wendell Piez [mailto:wapiez@xxxxxxxxxxxxxxxx]
Sent: Friday, October 31, 2008 1:44 PM
Subject: Re: [xsl] XSLT 2.1
At 01:21 PM 10/31/2008, you wrote:
>I know that my "true identity template" is far out but this is what we
>must do today to counter the limits of the "XQuery 1.0 and XPath 2.0
>Data Model" (XDM).
>But I would like to know if my common sense approach is right: if an
>XSLT stylesheet works in an XSLT processor then it could also have
>been part of the XSLT processor in the first place?
>That is, what I propose could have been done, but it would be a too
>clumsy road to follow?
I rather doubt that the "common sense" reasoning above is correct.
There is too much variation among possible implementation strategies
(now and in the future) to say generally that because one processor
does a thing, therefore any processor should be able to do it.
Consider the specification for a system that chops onions. The pieces
of onion must be uniform in size within certain parameters, say
between 0.5 square cm and 2 square cm. One system employs a knife and
cutting board to accomplish this successfully. Another works more
like a food processor: the onions are dropped in a bin, a rotary
knife blade is applied for a certain time, and the onions are chopped.
The manufacturer of the food processor also distributes a blender
attachment which can make fruit smoothies. It is easily done, but
nothing analogous can be done with the knife to accomplish the same end.
XSLT is not like either the food processor or the knife attachment:
it's the specification of the chopped onions. What is easily
accomplished by one XSLT (onion-chopping) implementation -- a fruit
smoothie -- is not necessarily accomplished by another XSLT
Asking that the literal form of XML input must be respected by an
XSLT engine and reproducible on output -- however that is defined to
include or exclude entity references, attribute order, whitespace in
tags, and other syntactic distinctions that are semantically opaque
in XML qua XML -- is liable to be much easier to accomplish using
some approaches than others, if not altogether impossible to do with
some that may be very good at XSLT in every other respect.
Note also that this has nothing to do with how reasonable the
requirement is to make fruit smoothies. If it's a reasonable
requirement, it can be specified, and a technology developed to do it.
Wendell Piez mailto:wapiez@xxxxxxxxxxxxxxxx
Mulberry Technologies, Inc. http://www.mulberrytech.com
17 West Jefferson Street Direct Phone: 301/315-9635
Suite 207 Phone: 301/315-9631
Rockville, MD 20850 Fax: 301/315-8285
Mulberry Technologies: A Consultancy Specializing in SGML and XML