[jira] [Resolved] (CAMEL-3927) CXF's Provider service using the payload data-format may not transfer the payload correctly

["William Tam (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CAMEL-3927?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

William Tam resolved CAMEL-3927.
--------------------------------

    Resolution: Fixed

Committed revision 1098630.  Thanks Aki Yoshida for the Jira and the patch.


> CXF's Provider service using the payload data-format may not transfer the payload correctly
> -------------------------------------------------------------------------------------------
>
>                 Key: CAMEL-3927
>                 URL: https://issues.apache.org/jira/browse/CAMEL-3927
>             Project: Camel
>          Issue Type: Bug
>          Components: camel-cxf
>    Affects Versions: 2.7.1
>            Reporter: Aki Yoshida
>            Assignee: William Tam
>             Fix For: 2.8.0
>
>         Attachments: camel-cxf-20110429b-patch.txt
>
>   Original Estimate: 2h
>  Remaining Estimate: 2h
>
> When using the Provider based service (i.e., one using the Provider<Source>'s invoke(Source m) signature), I observed that the payload content is not correctly extracted in the first call. The reason seems to be that the content is passed as a StaxSource instance to the camel's cxf binding component (DefaultCxfBinding) and this expects currently a DOMSource for the payload data format and therefore it not able to extract the payload content.
> Some more information is described in a related mail thread at camel-user 
> http://camel.465427.n5.nabble.com/Question-on-RemoveClassTypeInterceptor-in-Camel-Cxf-s-Payload-mode-td4358188.html
> This problem can be reproduced using CxfConsumerProviderTest when this test is executed using the payload data format. I suppose this test should actually be executed in the payload data format mode. But this is currently not the case.
> So, I am attaching the modified CxfConsumerProviderTest that uses the payload data format. This test will fail with the current trunk camel-cxf implementation. I am also attatching the patch file of the suggested change in DefaultCxfBinding that can resolve this issue and successfully pass this test.
> I think we can eventually consider a different CxfPayload implementation that does not require a DOM object at all. But for now, I think we can live with this correction to avoid this payload extraction issue.
> Thanks.
> Regards, Aki

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira