Just saw it.

Seems almost perfect, just wonder is managedbean shouldnt be checked
(cdi)...or if we shouldnt save this scanning if not done to avoid a long
field browsing
Le 1 sept. 2012 16:27, "Enrico Olivelli" <eolive...@gmail.com> a écrit :

> in the patch I attached a new test case in CheckPersistenceContextUsageTe*
> *st that uses that API
>
> Il 01/09/2012 16:25, Romain Manni-Bucau ha scritto:
>
>> Sounds not bad for me.
>>
>> Note: we have some cool api to test validations maybe you can enhance the
>> patch with such a test (rules package)
>> Le 1 sept. 2012 15:40, "Enrico Olivelli" <eolive...@gmail.com> a écrit :
>>
>>  I attached a patch (code + test) for this issue
>>> https://issues.apache.org/****jira/browse/OPENEJB-731<https://issues.apache.org/**jira/browse/OPENEJB-731>
>>> <https:**//issues.apache.org/jira/**browse/OPENEJB-731<https://issues.apache.org/jira/browse/OPENEJB-731>
>>> >
>>>
>>> It recognizes the most common usecase I think
>>> @PersistenceContext(unitName='****someRESOURCE_LOCALunit" EntityManager)
>>>
>>> does it work ?
>>>
>>> - Enrico
>>>
>>>
>>> Il 28/08/2012 05:59, David Blevins ha scritto:
>>>
>>>  On Aug 26, 2012, at 11:06 PM, Enrico Olivelli wrote:
>>>>
>>>>   I'm implementing https://issues.apache.org/****
>>>> jira/browse/OPENEJB-731<https://issues.apache.org/**jira/browse/OPENEJB-731>
>>>> <https:**//issues.apache.org/jira/**browse/OPENEJB-731<https://issues.apache.org/jira/browse/OPENEJB-731>
>>>> >
>>>>
>>>>> I'm adding it to:
>>>>> org.apache.openejb.config.****rules.CheckPersistenceRefs
>>>>>
>>>>> but I think I need the real PersistenceUnit, what is the best way to
>>>>> get
>>>>> it ?
>>>>>
>>>>>  The persistence-unit definitions will be in the AppModule.**
>>>> getPersistenceModules();
>>>>
>>>> Be warned that the actual matching of persistence units to references
>>>> can
>>>> get a little complicated.  There is actually a lesser known and lesser
>>>> used
>>>> URI syntax for identifying a jar and a unit name.  So we'll either need
>>>> to
>>>> be ok with giving up if we are unable to resolve it, or to duplicate the
>>>> resolution logic.
>>>>
>>>> What check were you imagining?  There's a check we don't have about
>>>> making sure an @EntityManager reference doesn't point to an
>>>> RESOURCE_LOCAL
>>>> unit.  Would be great to get that check, such a common mistake.
>>>>
>>>> I did a whole section on it at JAXConf:
>>>>
>>>> http://svn.apache.org/repos/****asf/openejb/trunk/**<http://svn.apache.org/repos/**asf/openejb/trunk/**>
>>>> presentations/2012_JAXConf_****Tomcat_to_JavaEE_with_TomEE.****pdf<
>>>> http://svn.apache.org/**repos/asf/openejb/trunk/**
>>>> presentations/2012_JAXConf_**Tomcat_to_JavaEE_with_TomEE.**pdf<http://svn.apache.org/repos/asf/openejb/trunk/presentations/2012_JAXConf_Tomcat_to_JavaEE_with_TomEE.pdf>
>>>> >
>>>>
>>>>
>>>> -David
>>>>
>>>>
>>>>
>

Reply via email to