I agree, returning null is perfectly valid and shall get treated the same as an empty Set. I created CDITCK-444
LieGrue, strub On Thursday, 11 September 2014, 23:00, Arne Limburg <[email protected]> wrote: > > >Hi Reini, > >I took a look at the test you mention and into the spec, too. >I guess, the test is dedicated to "Any bean which has its bean attributes >altered must have it¹s definition validated during deployment validation.", >but I agree with you, that the spec nowhere states that getStereotypes() >may not return null. > >So what to do with it? > >Write a mail to the cdi-dev list ([email protected]) and post your >question there. > >If no one can tell you, where in the spec is specified that >getStereotypes() may not return null, I¹ll open an issue for the tck at >(https://issues.jboss.org/browse/CDITCK) and we¹ll exclude that test until >the issue is resolved. >Btw. we should open a spec issue for CDI 2.0, too, to clarify, if >getQualifiers(), getStereotypes() and so on may return null. > >Cheers, >Arne > > > >Am 11.09.14 21:06 schrieb "Reinhard Sandtner" unter ><[email protected]>: > >>Hey guys, >> >>i try to fix >>org.jboss.cdi.tck.tests.extensions.lifecycle.processBeanAttributes.broken. >>invalid.InvalidStereotypeTest but i got no idea why a DefinitionException >>is expected. >> >>In >>org.jboss.cdi.tck.tests.extensions.lifecycle.processBeanAttributes.broken. >>invalid.InvalidStereotypeExtension they override getStereotypes() and >>return <null> (maybe they want to return the >>org.jboss.cdi.tck.tests.extensions.lifecycle.processBeanAttributes.broken. >>invalid.PlainOldAnnotation which is no stereotype) >> >>i¹m not sure if it¹s not allowed to return <null> in getStereotypes() >>(spec: 2.8 StereotypesŠ A bean may declare zero, one or multiple > >>stereotypes.) or if there is a bug in the tck? >> >>lg >>reini > > >
