> 
> when restricting a complex type, is it actually allowed (at 
> all) to change the type of an attribute defined at the 
> complex type to be restricted ?
> 

Yes, subject to the rules in
http://www.w3.org/TR/xmlschema-1/#derivation-ok-restriction

To paraphrase the rules:

2.1.1, 3 if there's an attribute that's required in the base type then =
it
must be present (and required) in the derived type

2.1.2 the type of the attribute in the derived type must be a =
restriction of
the type of the corresponding attribute in the base type

2.1.3 if the base type specifies a fixed value, then the derived type =
must
specify the same fixed value.

(Wouldn't it be nice if the spec were written like that, instead of =
saying
things like "what would have been the {name} and {target namespace} of =
the
{attribute declaration} of an attribute use in the set per clause 1 =
above
but for the =B7actual value=B7 of the use [attribute] of the relevant
<attribute> among the [children] of <restriction> being prohibited.")

Remember of course, that unlike child elements, attributes are =
automatically
inherited from the base type unless you explicitly exclude them (an
asymmetry I've always found very ugly).

Michael Kay
http://www.saxonica.com/


From debora.vanni@t... Fri Sep 22 08:49:02 2006
Received: from maggie.w