From: Adrian B. <adr...@jb...> - 2005-11-21 19:48:31
|
On Mon, 2005-11-21 at 14:14, Scott M Stark wrote: > So while I agree that this is a valid abstraction, its too abstract. > Looking at the configuration I have no idea that there is an "ENC > pattern" for which I can write a decorator to build a service like > JNDIView. In the absence of any pattern which categorizes like metadata, > the JNDIView decorator will simply be the externalization of the > implementation details that happen today be hardcoded into the JNDIView. > That is an improvement, but I think we need to go further in terms of > being able to "discover" which common metadata patterns like an ENC > exist so that for M patterns there are not M decorates that internally > have N bindings to the N services that somehow conform to the pattern. > I think you need to use concrete examples. *Your comment* is too abstract :-) What you said is not correct. There is no like metadata or multiple patterns. The pattern is in the decorator and it injects a standard piece of metadata into JNDIView. There is no pulling or discovery here. The only thing that requires any mutliples (short of the user actually requesting different behaviour in different places :-) is the mapping of deployment type specific metadata to our own model. ejb-jar.xml -> ENCMetaData web.xml -> ENCMetaData > ------------------------------------------------------- > This SF.Net email is sponsored by the JBoss Inc. Get Certified Today > Register for a JBoss Training Course. Free Certification Exam > for All Training Attendees Through End of 2005. For more info visit: > http://ads.osdn.com/?ad_id=7628&alloc_id=16845&op=click > _______________________________________________ > JBoss-Development mailing list > JBo...@li... > https://lists.sourceforge.net/lists/listinfo/jboss-development -- xxxxxxxxxxxxxxxxxxxxxxxx Adrian Brock Chief Scientist JBoss Inc. xxxxxxxxxxxxxxxxxxxxxxxx |