[ 
http://issues.apache.org/jira/browse/TUSCANY-195?page=comments#action_12376239 
] 

ant elder commented on TUSCANY-195:
-----------------------------------

This hack only works for externalServices, not for entryPoints. Clients can 
call SDOUtil.registerStaticTypes before calling the externalService, but I 
can't see how to do that for entryPoints.

The Tuscany runtime makes available the WS entryPoint, a WS request comes in 
and the XML needs to be converted to Java objects before the SCA componet is 
invoked. As nothing has registerged the SDO factory the result is a 
DynamicDataObject instead of the Objects the component is expecting. All the 
Axis2 binding code has to go on is the service interface class, and that has 
been dynamically gen'd from the WSDL so its not even the SEI class the 
component is using or even in the same class loader.

This is also why the TomcatIntegrationTestCases are broken with this hack 
applied.


> Load time EMF registering of URI to dynamic data object before static types 
> can be registered.
> ----------------------------------------------------------------------------------------------
>
>          Key: TUSCANY-195
>          URL: http://issues.apache.org/jira/browse/TUSCANY-195
>      Project: Tuscany
>         Type: Bug

>   Components: Java SCA Axis Binding
>  Environment: ALL current build
>     Reporter: Rick Rineholt
>     Priority: Blocker

>
> The loading of the TypeHelper in  
> ExternalWebServiceBuilder.createExternalWebServiceClient() line: 183 loads 
> into the EMF registry the URI that most likely maps to a static generated 
> type.  But this is occuring at load/build time prior to even any statically 
> generated application SDO's are in scope. Or before they can be registered by 
> the application.  This mapping is then done to DynamicDataObjects.  This 
> results in a class cast exception when the generated interface expects the 
> static types and a DDO is supplied.  My very limited understanding of EMF/SDO 
> is that once the mapping is in place that subsequent registering of static 
> types will not have any affect.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira

Reply via email to