openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Francesco Chicchiriccò <>
Subject Re: JPA 2.1 Progress
Date Fri, 13 Oct 2017 08:37:05 GMT
Hi Michael,
sorry for the late reply and thanks for your appreciation!

My 2c on the questions below, especially because I also have some 
interest in being able to run a non-SNAPSHOT version of OpenJPA 3.0.0 
with Syncope 2.1.0 / Spring 5.

AFAICT, the current trunk (e.g. 3.0.0-SNAPSHOT) already features some 
interesting enhancements when compared to 2.4.x (e.g. 2.4.3-SNAPSHOT), 
including Java 8 / JPA 2.1 compatibility, even though the implementation 
is not yet complete.

We also discussed [1] and decided [2] to move on with JPA 2.2 on trunk, 
and to a recent inquiry [3] Mark suggested that the possibility of 
releasing soon some kind of intermediate release is indeed possible.

Even though the final 3.0.0, fully JPA 2.2 is possibly far in the 
future, would it be realistic to consolidate what we have right now into 
a 3.0.0-M1 release?



On 11/10/2017 10:32, Michael Wiles wrote:
> Hi All
> As you know Spring 5 requires JEE7 as a minimum requirement.
> I am keen to upgrade to Spring 5 as I want to upgrade to Elastic search 5
> and do do that we need Spring Data 2 for that and for that we need Spring 5.
> Let me say that I can only recommend openjpa.
> I say this because in my endeavours to upgrade to Spring 5 I switched to
> using Hibernate and cursed hibernate because it required that every getter
> had a corresponding setter (even if those methods had nothing to do with
> persistence) unless the existing method is annotated with @Transient. This
> rather ludicrous requirement disqualified hibernate in my eyes.
> Then I tried to eclipselink and again my app did not start because the
> Spring Data meta model generating failed. Something to do with having
> Spring data repositories on composite keys (I have a lot of these) -
> notwithstanding the fact that I had to build my own version of eclipselink
> because I got a stackoverflow because I use an annotation that was
> annotated with itself (and they support meta annotations).
> What is interesting is that I used the 3.0.0-SNAPSHOT branch of Openjpa in
> my code and was able to run some of my integration tests successfully on
> Spring 5.
> So given all the above, I am unwilling to migrate to hiberate or
> eclipselink - this I do not want to do.
> I'm sorry to ask such a blunt question: When will a JPA 2.1 compliant
> version of openjpa be ready? I guess I would like some indication about
> whether it will weeks, months or years?
> In this regard, what can I do to help? I do not want to switch to hibernate
> or eclipselink so I would be willing to help where I can... I'm not sure
> I'd be able to do some of the more hard core stuff - although I do get up
> close and personal with the low level code from time to time - but I could
> help with testing and/or documentation... I do have a fairly robust
> application which does use a lot of the jpa specification.
> Thank you - and thanks again for what I think is the best JPA
> implementation around.
> Michael

Francesco Chicchiriccò

Tirasa - Open Source Excellence

Member at The Apache Software Foundation
Syncope, Cocoon, Olingo, CXF, OpenJPA, PonyMail

View raw message