I agree with you. The spec doesn't seem to provide any requirement that decorators should be injected directly into a) The injection point of the Decorated bean or b) each other.
I guess I'm sure sure how you can validate a TCK test with "secondary evidence" and the intentions of the EG. It seems to me that it needs to verify something that is clearly spelled out in the spec. I think the quote Eric pointed out certainly leaves some room for implementation choices. Sincerely, Joe On Mon, Aug 30, 2010 at 6:28 AM, Gurkan Erdogdu <[email protected]> wrote: > Seems that RedHat rejects TCK challaenge that I did open. Rejection comment is > given at https://jira.jboss.org/browse/CDITCK-137 > > Joe, how could we go with this ? WDYT? > > Thanks; > > --Gurkan > > > ________________________________ > From: Eric Covener <[email protected]> > To: [email protected] > Sent: Sun, August 29, 2010 3:42:12 PM > Subject: Re: CDI TCK Issues > > On Sun, Aug 29, 2010 at 3:08 AM, Gurkan Erdogdu <[email protected]> > wrote: >> Hello; >> >> We had problems with CDI-TCK 137 and CDI-TCK 138. I have put those issues >> into >> table but not responded so far. >> >> Please have a look and tell WDYT? >> >> https://jira.jboss.org/browse/CDITCK-137 >> https://jira.jboss.org/browse/CDITCK-138 > > The tests don't seem to relate to any requirement in the spec. IMO the > spec even alludes to the idea that the delegate injection points may > not be the next member of the chain: > > "The delegate object implements the delegate type and delegates method > invocations to remaining uninvoked decorators and eventually to the > bean" > > > -- > Eric Covener > [email protected] > > >
