Re: [weld-dev] [jsr-299-eg] Updated spec with @BeanTypes and clarification of @New
Jason T. Greene wrote:
Roberto Chinnici wrote:
> Gavin King wrote:
>>
>> You have provided no argumentation in support of "should not".
>>
>> Whereas I have provided a Very Good Reason why they *should* be
>> injectable, with limited semantics, using the special @New annotation.
>>
>
> OK, let me ask you a simpler question: if I have a web module, say,
> without a beans.xml
> descriptor, can I lookup a BeanManager in JNDI under
> java:comp/BeanManager? Currently
> the platform spec says you can't.
>
Are we all talking about the same thing? Earlier you mentioned use the
term "bean" instead of "class". Are you having an issue with the
terminology, or is it the capability you have a problem with?
To be clear, when you say @New blah, the resulting bean instance
belongs to the beanmanager associated with the injection point. The
"first" instance, if it even exists, is not touched in any way by 299.
So all we are talking about is a way to reference any class that fits
the rules of a 299 bean. In other words, anything which can be a 299
bean, has a "new" instance implicitly created. The source of that bean
is not really important, other than it has to be visible to the
classloader.
My issue is that I need to know at deployment time what components exist
in the application, so I can do the right thing wrt @Resource and friends.
The terminology is important because bean implies component. If you
called "a class that fits the rules of a 299 bean" a pre-bean, for
example, then I think I could reasonably excuse the deployment tool from
having to look at them.