Thanks, David for useful thoughts.
On Sun, Oct 3, 2010 at 3:15 PM, David Carlisle
<davidc(_at_)nag(_dot_)co(_dot_)uk> wrote:
no but they could both have been subtypes of a new numeric type sitting
between the existing ones and xs:atomic type.
I think xs:float cannot be a true subtype of xs:double as per rules of
value spaces of XML Schema (primitive) type system, because computing
values of double and float can have rounding errors.
but because it isn't really a type there have to be special XDM type
promotion rules.
I agree.
It seems to me that you kind of suggested something like following (a
fictitious XML Schema type hierarchy tree):
xs:anyAtomicType
-> myNumericType (base-type -> a combination of double, float)
-> double
-> float
So now may be as per above example, we can design (a custom-design) an
instance of "myNumericType" which is child of double and/or float.
--
Regards,
Mukul Gandhi
--~------------------------------------------------------------------
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
To unsubscribe, go to: http://lists.mulberrytech.com/xsl-list/
or e-mail: <mailto:xsl-list-unsubscribe(_at_)lists(_dot_)mulberrytech(_dot_)com>
--~--