Order relation of gMonth type

Dear XML Schema WG members,

First of all, congratulation for PR!


> This means that the specification is stable and that implementation
> experience has been gathered showing that each feature of the
> specification can be implemented. 

By the way, please tell me the name of the program that thoroughly
implemented part 2 of the spec. As far as I know, there is no reference
implementation.


I found a problem(?) in order-relation of gMonth type.


Let us compare two instances of gMonth type.

(P)  --02--Z            (March without time zone)
(Q)  --03---08:00       (March in Pacific Standard Time)

According to the spec (section 3.2.7.3),

(A) normalize P and Q.

   P is already normalized.
   Q has to be normalized, and that's where the problem happens.


According to the spec, addition must be done as defined in Appendix E of
the spec.

However, in appendix E is the following description.

> If a field in S is not specified, it is treated in the calculation as if
> it were the minimum allowed value in that field, however, after the
> calculation is concluded, the corresponding field in E is removed (set
> to unspecified). 

Because of this, this result has to be "--02--Z", although the intermediate
result of this addition is "28th,Feb 16:00:00Z".


The net result is, P==Q.

My question is, is this the intention of Schema WG?
It seems to me that this result is unintuitive, to say the least.


regards,
----------------------
K.Kawaguchi
E-Mail: k-kawa@bigfoot.com

Received on Monday, 19 March 2001 19:03:48 UTC