[weld-dev] archives containing a beans.xml with bean-discovery-mode=all

[Emily Jiang]

Hi Martin,
I totally agree with you. In integrating CDI 1.2 to WebSphere Liberty
profile, I interpreted the latter.

e.g. In a war with bean-discovery-mod=none, if one of its libs is a bean
archive, the cdi is enabled for the war module. The servlet in this war
will support injections.

However, it gives performance penalty, e.g. creating bda etc. I would like
to know what the other application servers (GlassFish or WildFly) behave.
It will be nice if the spec is clearer on this. I will raise a spec issue
on this.
Thanks
Emily

On Tue, Aug 18, 2015 at 2:36 PM, Martin Kouba <mkouba at redhat.com> wrote:

> Hi Emily,
>
> Dne 18.8.2015 v 11:36 Emily Jiang napsal(a):
>
>>  From CDI spec,
>> An archive which:
>> • contains a beans.xml file with the bean-discovery-mode of none, or,
>> • contains an extension and no beans.xml file
>> is not a bean archive.
>>
>> I think it is no need to create a bda for such bean archives, which
>> implies JavaEE component classes cannot support injection in these
>> archives. These archives are ignored basically. Right?
>>
>
> Yes, these archives are ignored from the CDI container point of view. As
> regards the injection support for Java EE component classes - this is not
> clearly defined. Java EE spec states in "EE.5.2.5
> Annotations and Injection":
>
> "The component classes listed in Table EE.5-1 with support level Standard
> all support Java EE resource injection, as well as PostConstruct and
> PreDestroy callbacks. In addition, if CDI is enabled—which it is by
> default—these classes also support CDI injection, as described in Section
> EE.5.24, Support for Dependency Injection, and the use of interceptors."
>
> However, it is not clear what "if CDI is enabled" means. One can argue
> that "CDI is enabled" if the component class resides in a bean archive. The
> other interpretation (the one I personally prefer) might be that "CDI is
> enabled" if a CDI container is initialized for the application (i.e.
> there's at least one CDI bean archive).
>
> I think we should check what Java EE RI (GlassFish) and other
> implementations (WildFly) do.
>
>
>> The spec does say an extension can exist in a non-bean archive but it
>> does not say anything about JavaEE component classes.
>>
>>
>> --
>> Thanks
>> Emily
>> =================
>> Emily Jiang
>> ejiang at apache.org <mailto:ejiang at apache.org>
>>
>>
>> _______________________________________________
>> weld-dev mailing list
>> weld-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/weld-dev
>>
>>
> --
> Martin Kouba
> Software Engineer
> Red Hat, Czech Republic
>



-- 
Thanks
Emily
=================
Emily Jiang
ejiang at apache.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/weld-dev/attachments/20150818/38beb5e9/attachment.html