Re: alternating tags in a list?

Subject: Re: alternating tags in a list?
From: Ray Cromwell <ray@xxxxxxxxxxxxxxx>
Date: Tue, 15 Dec 1998 15:58:17 -0500 (EST)
 I don't see the big deal about allowing script code to be evaluated
via XSL. If the XSL-WG isn't willing to add the neccessary features to
XSL to make it useful for most web developers, what's the point?

  You're suggesting that if XSL can't handle a particular query, that
someone post process with DOM+script. But processing nodes via DOM
in a language like Java is painful. Lots of getElementsByThisOrThat,
followed by for/while loops, following by accessor calls to 
Attributes trying to perform a match.

 In other words, the irritating, non-concise, bug-probable code,
that should be handled by a nice tailor made language using
a standard matching engine. XSL could alleviate most of
this work by allowing script-coders to use a "Visitor Pattern"
where XSL visits the nodes, and the script code gets ran on 
each node in the query.

  I used to think I understood what XSL was, what it was going to do
for me, but now I'm not so sure. I thought it was about style, and
rendering XML documents on the web. But apparently, that changed. Now
it seems it should be called XTL (extensible transformation language),
since now most of the talk is about transformation. But it doesn't
seem to be great for transformation either if it can't do simple
thinks like an even-odd rule.  What is it, a style language or
a transformation language? It doesn't seem to do either thing
particularly well right now.

  To top it off, there seems to be a resistence on the part of some
people to either 1) add the appropriate power to the matching language
or 2) temporarily break out to script code.  I think this is naive. It
is obvious to me that XSL lacks the power to do a lot of common web
formatting cases that is now handled by ASP code. This means there
will be a huge web developer demand for this functionality, and
commercial companies *will* add it to the language. It's not a
question of *if*, but when. Of course, there will be many
incompatibilities between these extensions, with one company
mostly dominating (likely MS) as a defacto standard, and finally, 
the W3C will be forced to bite the bullet and declare the
extensions part of the standard.

-Ray










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


Current Thread