W3C home > Mailing lists > Public > xmlp-comments@w3.org > March 2004

Round 5- 20040323 - SOAP v1.2(c) Test Collection

From: <cheekai@SoftML.Net>
Date: Mon, 22 Mar 2004 23:51:23 -0500 (EST)
Message-Id: <200403230449.i2N4nq8K008725@eastgate.starhub.net.sg>
To: XMLP WG <xmlp-comments@w3.org>
Cc: Siew Choo <siewchoo@i2r.a-star.edu.sg>, Ramasamy S/O Valliappan <rama@SIMTech.a-star.edu.sg>, Diwakar Jebaraj <diwakar@SIMTech.a-star.edu.sg>, ITSC Secretariat <nitsc@ida.gov.sg>



Hi,

A few more errors in the tests were found during a recent Web Services 
Plug Fest (WSPF) organized by ITSC Singapore.

Issues reported below were discovered together with the help and 
inputs from the following people:

Siew Choo <siewchoo@i2r.a-star.edu.sg>
Ramasamy S/O Valliappan <rama@SIMTech.a-star.edu.sg>
Diwakar Jebaraj <diwakar@SIMTech.a-star.edu.sg>

as well as the participating companies in the WSPF event.


Thanks.



Best Regards,
Chin Chee-Kai
SoftML
Tel: +65-6820-2979
Fax: +65-6743-7875
Email: cheekai@SoftML.Net
http://SoftML.Net/


--------------------------------------------------------------

Test:T53
Test:SBR1-echoDate (T96)

According to XML Schema, the type "xsd:dateTime" should be
normalized according to the REVERSE of the sign of the hours
indicated.  The current description performs directly the
arithmetic indicated instead of reversing them.

Strictly speaking, the test itself could spell out a test 
semantic that says, "in this test case, 'normalize' means
perform the arithmetic adjustments as manifested by the
unnormalized form", and there would be nothing "wrong"
about the test.

But the suggestion here is that it may be better to simply
follow what has been defined in XML Schema date-time normalization
as doing otherwise does not serve much of a point, and actually
tends to be confusing when the description does not indicate
a different way to normalize dateTime values from the way
specified by XML Schema.

Therefore, assuming the above correction of interpretation to
align with XML Schema's specification, Node C's response 
for both T53 and T96 should then be changed from:
+.+.+.+.+.+.+.+.+.+.+.+.+.+.
<return xsi:type="xsd:date">1956-10-18T15:20:00Z</return>
+.+.+.+.+.+.+.+.+.+.+.+.+.+.
to:
+.+.+.+.+.+.+.+.+.+.+.+.+.+.
<return xsi:type="xsd:date">1956-10-19T05:20:00Z</return>
+.+.+.+.+.+.+.+.+.+.+.+.+.+.

--------------------------------------------------------------
Received on Wednesday, 24 March 2004 09:22:52 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 23:40:22 UTC