Re: [hibernate-dev] Discussion of services

2010-10-12 Thread Max Rydahl Andersen
> > > Personally I think y'all are getting a little too hung up the on injection of > dependent services here. It is a quite limited scope and the implementation > is both dead simple and already done (its not like this is going to drag on > for years like we have seen elsewhere). It is not

Re: [hibernate-dev] Discussion of services

2010-10-07 Thread Steve Ebersole
On Monday, October 04, 2010 09:59:14 am Emmanuel Bernard wrote: > Not commenting on the idea of an Hibernate specific registry Good because the time for that was almost a year ago when I asked y'all to review the designs I was working on for 4 :) -- Steve Ebersole http://hibernate.org

Re: [hibernate-dev] Discussion of services

2010-10-07 Thread Steve Ebersole
On Monday, October 04, 2010 09:59:14 am Emmanuel Bernard wrote: > @InjectService > Why not JSR-303's @Inject with a qualifier if you really need some > additional metadata? I am not against the idea so much. The problems I had were twofold: 1) Personally I have no idea how to define these "quali

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Steve Ebersole
Ah, I had tried void and voic.class. Did not know about Void.class. Good to know. I will change that On Monday, October 04, 2010 09:59:14 am Emmanuel Bernard wrote: > Not commenting on the idea of an Hibernate specific registry, here are a > few comments on the design: > > Map should be Map

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Max Rydahl Andersen
I remember Mark P. mentioning they (drools) have some basic bootstrap for their core which is simplistic and then if containers want it can use another DI to set it up for user visibile customization (i.e. spring, cdi etc.) Not sure how they did it though - but might be worth a look ? /max On

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Emmanuel Bernard
Not commenting on the idea of an Hibernate specific registry, here are a few comments on the design: Map should be Map or whatever types will end up being used. public Class serviceRole() default InjectService.class; should probably be public Class serviceRole() default Void.class; Void.class

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Steve Ebersole
Not a simple one I am aware of. There is seam/weld/cdi which is more than we need. And I have been unable to find "real life" microcontainer use guides for something like this. Bear in mind that my bootstrap class is simply one choice/option to this. The services at the end of the day are ju

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Chris Bredesen
On 10/02/2010 06:39 PM, Steve Ebersole wrote: > https://sourceforge.net/apps/mediawiki/hibernate/index.php?title=Category:Services Isn't there a service-lifecycle-and-dependency-injection library in our arsenal we can use? -CB ___ hibernate-dev mailing

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Steve Ebersole
I wanted to make it explicit that this is not the initiator's type/role. Its the role/type of the service being initiated. I do not follow how you infer "get me the service initiated" to mean a method for doing the initiation. Given the vocabulary that has sort of grown, I'm ok with saying g

Re: [hibernate-dev] Discussion of services

2010-10-04 Thread Hardy Ferentschik
Nice. One comment, I think ServiceInitiator.getServiceInitiated() is a little confusing. Just by then name I thought it would be the method which initiates the service, but it rather just gives me the service type/name. Why not ServiceInitiator.getServiceType? --Hardy On Sun, 03 Oct 2010 00:

[hibernate-dev] Discussion of services

2010-10-02 Thread Steve Ebersole
https://sourceforge.net/apps/mediawiki/hibernate/index.php?title=Category:Services -- Steve Ebersole http://hibernate.org ___ hibernate-dev mailing list hibernate-dev@lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev