For your info, the change is quite minimal
https://github.com/emmanuelbernard/beanvalidation-spec/commit/f85703a2a39...
On 30 mai 2012, at 20:11, Emmanuel Bernard wrote:
It seems there is some kind of consensus that considering the runtime
type instead of the satin type to build the property path is better.
Check out
https://hibernate.onjira.com/browse/BVAL-226 and let me know if you object.
Otherwise, I plan to clarify this in the spec.
Emmanuel
_______________________________________________
beanvalidation-dev mailing list
beanvalidation-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/beanvalidation-dev