Christian Beikov

Can you please add a test case that reproducers this issue? I'd like to make sure we don't reintroduce such a regression again.

Hibernate ORM is in the process of certifying. I don't think there are any spec changes/fixes necessary, but we found some [problems in the TCK](https://github.com/jakartaee/platform-tck/pull/1287). There may be more, so...

Since we might request changes to the TCK, I would say that they "could" invalidate the CCR for EclipseLink.

Here are some further changes that improve the testing experience significantly: https://github.com/jakartaee/platform-tck/pull/1291

Thanks. We're working on it :) I hope we can finish this today since tomorrow is a holiday for most of our team

>It seems to me that we should either clearly maintain the distinction between `Predicate` and `BooleanExpression`, in which case logical operators don't apply to `BooleanExpression`s (or, if they do, they...

>I think that makes sense. It might actually be a mistake that Predicate is an Expression, but since it already is (ever since JPA 2), my prejudice is that we...

>Actually, the solution might be even simpler. Just add Comparable as an upper bound on the type parameter of NumberExpression. Does that work? >@beikov How about changing this to: >`public...

So you would rather have ```java entityManager.getMetamodel().entity(Book.class).getAttribute("owner").getJavaType() ``` return the concrete type i.e. `Owner.class` if `OwnerContainer` is only ever extended with specifying `Owner` as type variable, but start returning `Object.class`...

I guess you mean ```java entityManager.getMetamodel().entity(House.class).getAttribute("owner").getJavaType() ``` should return `HouseOwner`? This is where you seem to lack an understanding of how the metamodel is structured. An attribute is declared by...