- From: <noah_mendelsohn@us.ibm.com>
- Date: Fri, 18 May 2007 15:08:29 -0400
- To: "Pete Cordell" <petexmldev@tech-know-ware.com>
- Cc: "Rich Adili" <Rich.Adili@xata.com>, xmlschema-dev@w3.org
Pete Cordell writes: > In XSD schema, a base and the extra components of a derived > type are always > combined as if they are two components of an xs:sequence (i.e. base > components first, then derived components). Yes, in Schema 1.0 that's correct. The draft specifications for Schema 1.1 propose that <xsd:all>s be allowed to extend other <xsd:all>s, with the validation semantics being as if all the particles in the base and the derived were in a single <xsd:all> In other words, if the base allows ALL(A,B,C) and the extended allows ALL(D,E), then the resulting validation is as if the model were ALL(A,B,C,D,E), which allows each of the elements to appear once, but in any order. Noah -------------------------------------- Noah Mendelsohn IBM Corporation One Rogers Street Cambridge, MA 02142 1-617-693-4036 --------------------------------------
Received on Friday, 18 May 2007 21:24:54 UTC