From: Mark Struberg <struberg(a)yahoo.de&gt; To: cdi-dev <cdi-dev(a)lists.jboss.org&gt; Cc: Sent: Sunday, April 1, 2012 3:51 PM Subject: [cdi-dev] Alternative producer methods Dear EG colleagues! I have a question about @Alternative on producer methods: consider the following class: // here is NO alternative annotation! public class AlternativeOnProducerMethodOnlyBean {    @Produces  @Alternative    public IProducedBean generateBean2() {..} } So the class itself is NOT annotated as @Alternative I'm not sure if 5.1.1 allows this or the other way around: if this must be supported. Imo the important parts are (from 5.1.1) > Each child <class> element must specify the name of an alternative bean class. > If there is no class with the specified name, or if the class with the specified > name is not an alternative bean class, the container automatically detects > the problem and treats it as a deployment problem. The question is now how 'an alternative bean class' is to be interpreted ^^ The first paragraph of 5.1.1 says: >   An alternative is selected for the bean archive if either ... > * the alternative is a producer method, field or resource, and the > bean class that declares the method or field is listed, ... But does this qualify the AlternativeOnProducerMethodOnlyBean as 'alternative bean class'? Any ideas? Btw: the whole topic of disposal methods for @Alternative producers are completely unspecified... LieGrue, strub _______________________________________________ cdi-dev mailing list cdi-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/cdi-dev

I *think* that the expected behaviour in this case should be that the original disposer is disabled and only a disposer defined on the same bean class as the alternative producer should apply. If this is a correct interpretation of the intent, it needs to make its way into the spec. On 2012-04-01, at 12:06 PM, Mark Struberg wrote: > To be more clear what I mean with the @Disposal issue please consider the following class: > > public class DefaultBeanProducer > { > public static boolean gotDumped = false; > > @Produces @QualifierProducerBased > public IProducedBean generateBean() > { > return new ProducedBean("default", this); > } > > public void dumpBean(@Disposes @QualifierProducerBased IProducedBean bean) > { > gotDumped = true; > } > } > > > And the following @Alternative: > > @Alternative > public class AlternativeOnClassAndProducerMethodBean > { > @Produces > @Alternative > @QualifierProducerBased > public IProducedBean generateBean2() > { > return new ProducedBean("AlternativeOnClassAndProducerMethodBean", this); > } > } > > > > In OWB this currently creates the following Exception: > WebBeansConfigurationException: Producer method component of the disposal method : dumpBean in class : org.apache.webbeans.newtests.concepts.alternatives.common.DefaultBeanProducer must be in the same class! > > This behaviour is exactly as defined in the current spec afaik. > But certainly this is not very useful... > > wdyt? Did I overlook something? > > LieGrue, > strub > > > > > ----- Original Message ----- >> From: Mark Struberg <struberg(a)yahoo.de&gt; >> To: cdi-dev <cdi-dev(a)lists.jboss.org&gt; >> Cc: >> Sent: Sunday, April 1, 2012 3:51 PM >> Subject: [cdi-dev] Alternative producer methods >> >> Dear EG colleagues! >> >> >> I have a question about @Alternative on producer methods: >> consider the following class: >> >> // here is NO alternative annotation! >> >> public class AlternativeOnProducerMethodOnlyBean { >> @Produces @Alternative >> public IProducedBean generateBean2() {..} >> } >> >> So the class itself is NOT annotated as @Alternative >> >> I'm not sure if 5.1.1 allows this or the other way around: if this must be >> supported. >> >> Imo the important parts are (from 5.1.1) >>> Each child <class> element must specify the name of an alternative >> bean class. >> >>> If there is no class with the specified name, or if the class with the >> specified >> >>> name is not an alternative bean class, the container automatically detects >> >>> the problem and treats it as a deployment problem. >> >> The question is now how 'an alternative bean class' is to be interpreted >> ^^ >> The first paragraph of 5.1.1 says: >> >>> An alternative is selected for the bean archive if either ... >>> * the alternative is a producer method, field or resource, and the >> >>> bean class that declares the method or field is listed, ... >> >> >> But does this qualify the AlternativeOnProducerMethodOnlyBean as >> 'alternative bean class'? >> >> Any ideas? >> >> Btw: the whole topic of disposal methods for @Alternative producers are >> completely unspecified... >> >> LieGrue, >> strub >> >> _______________________________________________ >> cdi-dev mailing list >> cdi-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/cdi-dev >> > > _______________________________________________ > cdi-dev mailing list > cdi-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/cdi-dev _______________________________________________ cdi-dev mailing list cdi-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/cdi-dev

From: Pete Muir <pmuir(a)redhat.com&gt; To: Mark Struberg <struberg(a)yahoo.de&gt; Cc: cdi-dev <cdi-dev(a)lists.jboss.org&gt; Sent: Monday, April 2, 2012 11:30 AM Subject: Re: [cdi-dev] Alternative producer methods Right now, I can't find a better path through the spec to explain this, if anyone knows, then say, if not we should clarify for CDI 1.1 Looking at 11.1 getBeanClass() you can see that the bean class of a producer method (or field) is the class of the bean declaring the producer method (or field). This should probably be actually specified, rather than just in a interface description! From this, we can use Mark's reference to 5.1.1 to see that when we want to enable an alternative producer, we in fact use the bean class name. On 1 Apr 2012, at 14:51, Mark Struberg wrote: > Dear EG colleagues! > > > I have a question about @Alternative on producer methods: > consider the following class: > > // here is NO alternative annotation! > > public class AlternativeOnProducerMethodOnlyBean { >     @Produces  @Alternative >     public IProducedBean generateBean2() {..} > } > > So the class itself is NOT annotated as @Alternative > > I'm not sure if 5.1.1 allows this or the other way around: if this must be supported. > > Imo the important parts are (from 5.1.1) >> Each child <class> element must specify the name of an alternative bean class. > >> If there is no class with the specified name, or if the class with the specified > >> name is not an alternative bean class, the container automatically detects > >> the problem and treats it as a deployment problem. > > The question is now how 'an alternative bean class' is to be interpreted ^^ > The first paragraph of 5.1.1 says: > >>   An alternative is selected for the bean archive if either ... >> * the alternative is a producer method, field or resource, and the > >> bean class that declares the method or field is listed, ... > > > But does this qualify the AlternativeOnProducerMethodOnlyBean as 'alternative bean class'? > > Any ideas? > > Btw: the whole topic of disposal methods for @Alternative producers are completely unspecified... > > LieGrue, > strub > > _______________________________________________ > cdi-dev mailing list > cdi-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/cdi-dev