Steve, I do believe we are on the same page and there is no disagreement at
all. I think you understand my argument clearly and I do understand your
concerns regarding users.
On Wed, Nov 3, 2010 at 4:49 PM, Steve Ebersole <steve(a)hibernate.org> wrote:
On Wednesday, November 03, 2010, at 11:33 am, Tomasz Blachowicz
wrote:
> Steve, I think I know where you coming from. And my request is to do it
> *only* in the context of JPA meta-model. I'm far from making any changes
to
> the internals of Hibernate.
> My pain point is that if EntityType is not providing me with the
> @Entity.name I don't have any other means of getting that piece of
> information through meta-model other than getting the entity class by
name
> and checking @Entity in my code, which I'd consider pretty nasty hack.
> What's more FQN of entity class can be obtained not only by calling
> EntityType#getName, but also by calling EntityType#getJavaType (inherited
> from Type) or EntityType#getBindableJavaType (inherited form Bindable).
I already stated that I empathize with what you propose for these exact
reasons. I specifically said it was mainly because the JPA metamodel gives
you no other way to get this information. So not sure where the
disagreement
is.
I'm not talking about effects of these changes internally. Hibernate
itself
never really uses this value in relation to the entity (it registers the
import, that's it). Again, my concern is users. If you are using and
relying
on EntiytType.getName() to return the FQN you are screwed if you update to
whatever includes this change. At the very least the change should not
come
during a minor release rev. Now, if people don't really use this then its
not
such a big deal. Again outside my realm of experience, hence I ask...
---
Steve Ebersole <steve(a)hibernate.org>
http://hibernate.org
_______________________________________________
hibernate-dev mailing list
hibernate-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/hibernate-dev