[
Lists Home |
Date Index |
Thread Index
]
I wrote:
> I think that it would be better to have a consistent approach to
> partially ordered data types. One of:
>
> - creating subtypes that are ordered
> - using three-valued logic
> - performing an arbitrary conversion to give total ordering
but thinking about it, three-valued logic would be hard to do in XPath
2.0, given the requirements of:
- backwards compatibility with XPath 1.0 (which treats an empty node
set as false)
- compatibility with W3C XML Schema data types and therefore
xs:boolean, which only has two values: true and false (I guess
since I'm arguing against changing the semantics of the W3C XML
Schema data types I shouldn't suggest introducing a new xf:boolean
primitive data type)
I wonder whether an approach similar to that used for comparing
strings (i.e. allowing users to specify a particular collation to use)
would be a reasonable course of action.
Cheers,
Jeni
---
Jeni Tennison
http://www.jenitennison.com/
|
