Re: Whitespace normalization for union types

On Jun 1, 2005, at 6:00 PM, Kasimier Buchcik wrote:
> Hmm, I repeat [1]:
> "For all datatypes ·derived· by ·union·  whiteSpace does not apply
> directly; however, the normalization behavior of ·union· types is
> controlled by the value of whiteSpace on that one of the ·memberTypes·
> against which the ·union· is successfully validated."

I don't understand the last use of the word "union" in the quoted text. 
You seem to be reading the word to mean something like "intial value" 
or "pre-normalized string", which sounds reasonable. While I can't make 
sense of it literally, I was reading it to imply that for the member 
type to control the whiteSpace normalization, the value had to also be 
valid against the entire rest of the union type (including facets).

Now that I think about the ramifications, I like your interpretation 
better, mainly because it keeps the base type validation separate.

> I really need some definitive clarification on this.

I think the application of facets to unions is underspecified. 
Afterall, there are statements like:

     The value space and lexical space of a union datatype are the
     union of the value spaces and lexical spaces of its memberTypes.

which ignores the possibility of facets. I don't see any real 
improvement in the 1.1 working draft, either.

xan