Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-16 Thread Steve Ebersole
I found them your side of the conversation in my inbox :) Yes they were comments made on the fork branch which no longer exists :( The discussion centered around cases where HEM defines the same operations differently from CORE. Here it is the notion of persist(); the difference is that CORE

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-16 Thread Hardy Ferentschik
On 16 Jan 2012, at 12:42 AM, Steve Ebersole wrote: > In fact splitting those purposes alone solves the problem with > org.hibernate.engine.spi.JpaCascadingAction I believe > > As for org.hibernate.engine.spi.JpaCascadeStyle, TBH, I am not sure > exactly why we need this split between persist()

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-16 Thread Steve Ebersole
BTW, on master commons-annotations is in fact still needed so yes we need to account for it. However, just a heads up that on the metamodel branch, we no longer use commons-annotations. Some background that I alluded to before... Master is 4.1 + (ongoing) JPA 2.1 support. There is also a sha

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-16 Thread Steve Ebersole
Guys, lets move this discussion to the Jira. Thats the best way to capture all these points... On Thu 16 Aug 2012 09:25:58 AM CDT, Scott Marlow wrote: > Does the Bundle-Activator really belong in the Hibernate core project? > Or should it be in the external OSGI container? Or perhaps someone >

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-16 Thread Scott Marlow
Does the Bundle-Activator really belong in the Hibernate core project? Or should it be in the external OSGI container? Or perhaps someone may want to maintain an external project that generates the "ready to be used" Hibernate Bundle-Activator. On 08/16/2012 05:58 AM, mail...@bibbernet.org wr

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-16 Thread mailing
Hi Steve, > Not being well-versed in OSGi, could you explain what 'blueprint.xml' > is? > I guess the purpose is the same/similar to > /META-INF/services/javax.persistence.spi.PersistenceProvider defined by > JPA, yet OSGi specific? Ok, to give some background information, I will cite a section o

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-15 Thread Steve Ebersole
Ever hit send and immediately know you wanted to stop it/take it back ;) https://hibernate.onjira.com/browse/HHH-7527 On Wed 15 Aug 2012 07:39:29 PM CDT, Steve Ebersole wrote: > Martin, Brett, > > I went ahead and created a Jira for this. Lets discuss there. > > On Wed 15 Aug 2012 07:30:51 PM C

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-15 Thread Steve Ebersole
Martin, Brett, I went ahead and created a Jira for this. Lets discuss there. On Wed 15 Aug 2012 07:30:51 PM CDT, Brett Meyer wrote: > > Martin makes good points. I'd highly recommend moving this into JIRA. > As frequently as Hibernate gets used within OSGi contexts, following > some basic guidel

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-15 Thread Brett Meyer
Martin makes good points. I'd highly recommend moving this into JIRA. As frequently as Hibernate gets used within OSGi contexts, following some basic guidelines would be incredibly helpful. +1 for enforcing package splits, as well as ensuring the manifests are generated correctly/automaticall

Re: [hibernate-dev] OSGI manifests for hibernate-orm

2012-08-15 Thread Steve Ebersole
Martin, Totally agree about the org.hibernate.engine.spi bits. Thats always been fugly hack. Both org.hibernate.engine.spi.CascadeStyle and org.hibernate.engine.spi.CascadingAction attempt to combine contracts and enumerations, which I have never been a fan of here. I think its fine when tha