[xsl] The Perils of Sudden Type-Safety in XPath 2.0

Subject: [xsl] The Perils of Sudden Type-Safety in XPath 2.0
From: Gunther Schadow <gunther@xxxxxxxxxxxxxxxxxxxxxx>
Date: Tue, 18 Feb 2003 12:35:55 -0500
Hi, I am upgrading my stuff to Saxon 7.4 and I am biting my behind
now that I even started this (one day before a demo!!!).

Apparently now everything wants to be type safe and type rules
are being enforced according to XPath 2.0.

I do see the point for type safety in languages that never did
anything else, like Java. But I think suddenly making XPath 2.0
into a strongly typed language was a BAD, BAD, BAD idea!

This suddenly breaks a lot of existing XSLT code and even breaks
implementations that try to comply (such a Saxon) in aread that
where hitherto never a problem. I now get Saxon right out crashing
from ClassCastExceptions and stuff.

Please, please, can't this decision for XPath not turned around?
Could strong typing not be made optional? Why should one go through
the hassle of adding explicit type conversions if they do nothing
else than making the hitherto conveniently implicit conversions
explicit. What's the point of this?

Also what I don't like is that this now makes me drag reference
to W3C XML-Schema into my transforms everywhere just so I can
use the xs:integer() conversion, when I have no business with
W3C schema otherwise?

I am quite taken aback by this.

Gunther Schadow, M.D., Ph.D.                    gschadow@xxxxxxxxxxxxxxx
Medical Information Scientist      Regenstrief Institute for Health Care
Adjunct Assistant Professor        Indiana University School of Medicine
tel:1(317)630-7960                         http://aurora.regenstrief.org

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

Current Thread