I really don't like the idea of this being EE wide. CDI does a lot of
things that would perform terribly if you expanded the scope beyond the
application boundary. So we must be very careful about how any expansion
of these features is specified.
Even at the application level this is an expensive feature. It makes it
impossible for the app server to do any kind of offline indexing, or any
kind of caching of the annotation data, since a portable extension's
actions are not known to it. We can mitigate this to some extent by
requiring the extension to declare what exactly it touches via some kind
of configuration metadata. However, the problem here is still that we
are optimizing the case where the feature *doesn't* happen. I have
always thought that the case of adding support for new annotations was
much more common than modifying existing ones, so I still wonder how
valuable this behavior really is.
On 10/24/11 4:17 AM, Werner Keil wrote:
> +1
>
> On Mon, Oct 24, 2011 at 5:13 AM, Reza Rahman <reza_rahman_at_lycos.com
> <mailto:reza_rahman_at_lycos.com>> wrote:
>
> Pete has it right -- that's exactly what I meant. I don't see a
> reason for a performance issue.
>
> On 10/23/2011 1:07 PM, Pete Muir wrote:
>
> Sorry for the delay on this, I thought my messages were getting
> through but on talking to Linda, it appears they never arrived ;-)
>
> On 18 Oct 2011, at 00:25, Bill Shannon wrote:
>
> * Redefining Servlets, etc as managed beans -- the most
> tantalizing
> benefits to this is being able to use the CDI SPI in
> Servlets and using
> @Alternative/_at_Interceptor throughout all Java EE components.
>
> You should already be able to use the CDI SPI in Servlets,
> right?
> What prevents that currently?
>
> Reza is proposing that the CDI SPI, which can be used to alter
> the annotations seen on classes, methods, fields, parameters and
> constructors could be honoured by other Java EE components - IOW
> these other components would see the annotations as altered by a
> user application.
>
> This is a very powerful idea, but there are concerns from our
> team about the performance implications. Jason has more details
> here.
>
>
>
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 2012.0.1831 / Virus Database: 2092/4567 - Release Date:
> 10/22/11
>
>
>
>
>
>
>
> --
>
> Werner Keil | JCP Executive Committee Member (SE/EE) | Eclipse UOMo Lead
>
> Twitter @wernerkeil |Skype: werner.keil | www.eclipse.org/uomo
> <http://www.eclipse.org/uomo> | #EclipseUOMo
>
>
> * Java2Days: November 3-4 2011, Sofia, Bulgaria. Werner Keil, JCP EC
> Member, Agile Coach, UOMo Leadwill present "JSR 321: Trusted Java API,
> Enterprise Social Networking, Java EE 7 in the Cloud"
>
> * DevoXX: November 16 2011, Antwerp, Belgium. Werner Keil, JCP EC
> Member, Agile Coach, UOMo Lead will present "Java EE 7 Gathering"
>
--
Jason T. Greene
JBoss AS Lead / EAP Platform Architect
JBoss, a division of Red Hat