On 18 Nov 2010, at 16:50, Scott Ferguson wrote:
Pete Muir wrote:
> Hi Scott
>
> From the spec,
>
> "The event object must be of type
javax.enterprise.inject.spi.ProcessObserverMethod<T, X>, where T is the bean
> class of the bean that declares the observer method and X is the observed event type
of the observer method.
>
> public interface ProcessObserverMethod<T, X> {
> public AnnotatedParameter<T> getAnnotatedEventParameter();
> public ObserverMethod<X> getObserverMethod();
> public void addDefinitionError(Throwable t);
> }
> "
>
> It looks like this has got completely wrong against the javadoc and signatures
released with the spec, I will go to the EE CTS guys and see what the best path forward
is, however I suspect it is unfixable until we do a MR. I will also go back in and fix
this test so that it works correctly against the API as defined in the javadoc.
>
Thanks Pete.
I'd checked against the latest API and Weld but both were using the
javadoc not the spec, so I wasn't sure which way the changes were going.
So it's basically the same issue as the IF_EXISTS attribute name.
Right, and the Java EE CTS team at Oracle were very clear that the javadoc takes
precedence over the spec in this case (which I think is totally the wrong approach, for me
the spec is canonical).
Scott, could you file a CDI issue so that we don't loose track of this?
Thanks, and apologies for the screw up!
Pete