I agree that option two is a no go. Personally, I would like to keep the
whole product
range aligned in version number. Maybe we should go for 3.2 or 3.x for the
greater good :)
--Hardy
On Tue, 27 Jan 2009 18:09:33 +0100, Emmanuel Bernard
wrote:
The next version of Hibernate Validator w
I've written a small utility library that provides methods to work
with Types, specifically it contains such a method isAssignable(Type
supertype, Type type) that should satisfy your requirements.
I've been considering open-sourcing it so would be happy to do so if
you were interested?
Cheers,
M
If I can't find an other alternative that would surely be of help yes :)
Do you know how much of the subtype rules you have implemented?
The one that sounded non trivial to me were around detecting that
generic are supertypes of others:
- containment and equivalence
- capture conversion
-
I haven't gone into supporting type variables, hence I don't perform
type capturing or inference yet, although I would be interested in
adding this. For an example of what is currently implemented, the
following are from the tests:
Number <: Integer
Object <: Integer
List <: List
Collection <: Li
Latest refinements and batch of questions: 1 I gave up type subclassing, too much work and nobody seems excited 2 should we have and or should be directly under ? 3 should we have or should be under directly? 4 should we allow 30 with the risk of people doing 30blahor should we mandate the
Cool, that's already a nice start.
Let me see if I find something.
Would you be OK to contribute it under the ASL 2.0 license to us?
On Jan 29, 2009, at 09:44, Mark Hobson wrote:
I haven't gone into supporting type variables, hence I don't perform
type capturing or inference yet, although I w
ahem
excerpt from the Harmony code
package java.lang;
public final class Class implements Serializable, AnnotatedElement,
GenericDeclaration, Type {
[...]
public boolean isAssignableFrom(Class cls) {
return false;
}
}
I don't think they have what I am looking f
