I know I am asking way too much for someone to actually look at some code
;) but it would be amazing if someone versed in Classmate and/or
parameterized types could take a look at my implementation. I isolated
most of the code
in org.hibernate.boot.internal.AttributeConverterDescriptorImpl now and
in org.hibernate.boot.internal.AttributeConverterManager. I basically
completely re-wrote how auto-applicable converters are matched so that the
logic is centralized in these classe (local @Convert application is a
different beast).
On Mon, Oct 5, 2015 at 11:32 AM Hardy Ferentschik <hardy(a)hibernate.org>
wrote:
On Mon, Oct 05, 2015 at 02:03:54PM +0000, Steve Ebersole wrote:
> Bueller..?
>
> Any thoughts?
I am not familiar of the whole concept of AttributeConverter, but if you
need
full type resolution of a generic type, ClassMate is the way to go.
That's what we use in Hibernate Validator and that is also what I used when
working on the metamodel changes (former ORM 5) in order to
simplify/refactor
the annotation processing code.
--Hardy
> On Fri, Oct 2, 2015 at 6:59 PM Steve Ebersole <steve(a)hibernate.org>
wrote:
>
> > Reference HHH-10050[1]. There is a request to incorporate
parameterized
> > type resolution into the decision of whether an auto-apply
> > AttributeConverter should apply to a given attribute.
> >
> > For those on the list better at parameter type resolution, is the best
> > solution here to use ClassMate? Or is there a better way to achieve
this?
> >
> > [1]
https://hibernate.atlassian.net/browse/HHH-10050
> >
> _______________________________________________
> hibernate-dev mailing list
> hibernate-dev(a)lists.jboss.org
>
https://lists.jboss.org/mailman/listinfo/hibernate-dev