W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > November 2002

Re: TEST: datatype testcase

From: Jos De_Roo <jos.deroo.jd@belgium.agfa.com>
Date: Tue, 19 Nov 2002 00:55:43 +0100
To: "Jan Grant <Jan.Grant" <Jan.Grant@bristol.ac.uk>
Cc: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
Message-ID: <OF12555A91.1DAC9B57-ONC1256C75.0080324D-C1256C75.008381FF@agfa.be>

I now think that

  :john :birthday "1956-01-10"^^xsd:date .

RDFS/XSD entails

  :john :birthday "1956-01-10T+02:00"^^xsd:date .

is wrong

in http://www.w3.org/TR/xmlschema-2/ we can see that

If the time zone is included, both hours and minutes
must be present.

Since the lexical representation allows an optional
time zone indicator, date values are partially ordered
because it may not be possible to unequivocally
determine the order of two values one of which has
a time zone and the other does not.
If date values are considered as periods of time, the
order relation on date values is the order relation
on their starting instants. This is discussed in Order
relation on dateTime ( See also Adding
durations to dateTimes (E). Pairs of date values with
or without time zone indicators are totally ordered.

I was believing the entailment because Xerces's
new DateDatatypeValidator().compare(s1, s2)
gives equal, but I now see that .NET using System.Xml
DateTime.Compare(XmlConvert.ToDateTime(s1), XmlConvert.ToDateTime(s2))
does't give equal

-- ,
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/
Received on Monday, 18 November 2002 18:56:22 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:24:18 UTC