RE: restrictions with nested simpleTypes.

> I don't see this as being a violation. The spec does *not* say 'if
> memberTypes is non-empty there cannot be a simpleType child'. It
> merely says
> that if memberTypes is empty there MUST be at least one simpleType child.
> Does this make sense?

You're right.

I was misled by the parenthetical note in Section 2.5.1.3 which states:

"(that is, the order of the <simpleType> children of the <union> element, or
the order of the QNames in the memberTypes attribute)"

Notice the word "or". It didn't state that the simpleType children and
QNames should be concatenated like Section 5.1.3 does:

"The type definitions resolved to by the items in the value of the
memberTypes  [attribute], if any, in order, followed by the type definitions
corresponding to the simpleType  [children], if any, in order."

Jason.

Received on Monday, 11 December 2000 01:32:30 UTC