[ 
https://issues.apache.org/jira/browse/OWB-645?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14327199#comment-14327199
 ] 

Mark Struberg commented on OWB-645:
-----------------------------------

This is a totally different use case than you have. It has nothing to do with 
your issue. I will comment on your issue OWB-1038.

> InjectionPoint is null when using @Produces. NullPointerException caused by 
> CDI container providing a null InjectionPoint into a producer method.
> -------------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: OWB-645
>                 URL: https://issues.apache.org/jira/browse/OWB-645
>             Project: OpenWebBeans
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 1.0.0, 1.1.0, 1.1.2, 1.1.3
>         Environment: any platform 
>            Reporter: Rohit Dilip Kelapure
>            Assignee: Rohit Dilip Kelapure
>              Labels: InjectionPoint
>             Fix For: 1.1.4
>
>   Original Estimate: 72h
>  Remaining Estimate: 72h
>
> The CDI container must provide a bean with scope @Dependent, bean type 
> InjectionPoint and qualifier @Default, allowing dependent beans to obtain 
> information about the injection point
> to which they belong. In certain circumstances this InjectionPoint is null or 
> returns the wrong information about the injection point to which they belong.
> This is a two part problem
> 1. A bean with scope @Dependent may inject an instance of InjectionPoint and 
> access metadata relating to the injection point to which it belongs. This 
> instance of InjectionPoint is
> null when the @Inject of the dependent bean is in an EJB. 
> 2. CDI container uses ThreadLocals to track the current InjectionPoint. When 
> multiple injection points are interleaved, the CDI container does not 
> correctly track the state of of the
> InjectionPoint's such that sometimes the most deeply nested InjectionPoint 
> would still be on the thread local when it shouldn't be.
> The fix to the second problem is remedied by keeping a stack of injection 
> points. 
> The fix has MAY now  cause the InjectionPoint to be  null in cases where 
> earlier we wouldn't have before.
> CDI container needs to be  fixed to handle InjectionPoint injections in 
> Producer methods and interleaved injections.
> See Reference: http://www-01.ibm.com/support/docview.wss?uid=swg1PM51802



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to