Re: Datatype and identity constraints proposal of the day (17 May)

[James Clark <jjc@thaiopensource.com>]

> Implementations of datatype libraries *can* adopt order-independent,
> named parameters.  On the other hand, should RELAX NG care?  In other
> words, when two parameters have the same name, who should complain?
> The RELAX NG processor or implementations of datatype libraies?

Right.  That's exactly the issue.

> I  think that JamesC is proposing the latter.

I'm asking rather than proposing.

Another consideration is that XML Schema allows multiple "pattern" facets.

NOTE: It is a consequence of the schema representation constraint Multiple
patterns (§4.3.4.3) and of the rules for ·restriction· that ·pattern· facets
specified on the same step in a type derivation are ORed together, while
·pattern· facets specified on different steps of a type derivation are ANDed
together.

So this raises the issue: if you have multiple pattern params, should they
be ANDed or ORed?

Unless we allow multiple pattern params and AND them together, we will lose
functionality compared to XML Schema, and won't be able to represent all XML
Schema datatypes.

James