W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > July to September 2000

Re: X3D-related comments on Schema datatypes (LC144 and LC145)

From: Priscilla Walmsley <priscilla.walmsley@xmls.com>
Date: Fri, 29 Sep 2000 09:23:39 -0400
To: <brutzman@nps.navy.mil>
Cc: <www-xml-schema-comments@w3.org>
Message-ID: <002401c02a18$877ad650$838f143f@priscillaw01>
Hello Don,

Thank you for your comments on XML Schema, which were recorded in the Last
Call Issues List as LC-144 and LC-145.  The XML Schema working group has
asked me to reply to your message.

On your point number 2, as you know you can specify an allowed length for a
list, but not the tuple-ordinality of the list.   We considered this
request, but determined that we would not be able to add it in the first
version of XML Schema.

On point number 3, if I understand your example correctly, you would like to
constrain a list of floats so that every third  value is 1, e.g.
<normalVectors>0 0 1 1 2 1 0 0 1</normalVectors>  Unfortunately, the first
version of XML Schema will not allow you to constrain a list of floats in
this way.  All individual items in the list must be of the same type, and
therefore all constraints on the items must be the same. A
regular-expression pattern cannot be used to express this because patterns
cannot be applied to lists. (They can apply to every item in the list, but
not to the list as a whole.)

It seems that your choices are to (1) not validate based on these
constraints - if performance is a big issue - or (2) wrap each number in its
own tags, which would allow you the full power of XML Schema.

Please feel free to respond with any further questions or concerns on this
topic; we certainly appreciate your feedback.

Thank you,
Priscilla Walmsley
VP, Research
XMLSolutions Corporation
7929 Westpark Drive Suite 100
McLean, VA 22102
(877) XML-XMLS
Received on Friday, 29 September 2000 09:22:30 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:49:53 UTC