RE: question about subtyping and type annotations

I agree with your assessment that xs:integer(xs:byte(100)) should have a
type annotation of xs:integer. Mike apparently disagrees. You should
file a Mozilla request to add this to the test suite.

Best regards
Michael=20

> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On=20
> Behalf Of Per Bothner
> Sent: Thursday, February 23, 2006 7:30 PM
> To: [email protected]
> Subject: question about subtyping and type annotations
>=20
>=20
> I'm unclear on the semantic model for type annotations, and I can't
> find a clear answer in the specifications or in the test suite.
>=20
> I gather (and Saxon agrees) that:
>    100 instance of xs:byte =3D=3D> false
> because 100 has "type annotation" xs:integer, which is not a sub-type
> of xs:byte.
>=20
> Using the cast xs:byte creates a value with type annotation xs:byte:
>    xs:byte(100) instance of xs:byte =3D=3D> true
>=20
> What about xs:integer(xs:byte(100))?  I would expect this to return
> a value with type annotation xs:integer, but Saxon (b8-6-1) disagrees:
>    xs:integer(xs:byte(100)) instance of xs:byte =3D=3D> true ????
>=20
> This implies that if the argument already conforms to the=20
> casted-to-type
> then it it returned changed.  That is certainly a possible behavior,
> but another behavior is that casting to xs:integer returns a value
> with annotation xs:integer.  That seems more consistent to me.
>=20
> The formal semantics just refers to chapter 17 of Functions and
> Operators, which mainly specifies which conversions are valid.
> The formal semantics should probably specify the resulting "formal
> values" after a cast.
>=20
> Either way, this should be tested in the test suite.  I didn't
> find it (using a quick grep).
> --=20
> 	--Per Bothner
> [email protected]   http://per.bothner.com/
>=20
>=20