[ 
https://issues.apache.org/jira/browse/TUSCANY-2189?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Scott Kurz updated TUSCANY-2189:
--------------------------------

    Attachment: test.2189.jar

Sorry this doesn't exactly match the SCDL pasted it the JIRA text but it's the 
same idea

> Problems building the wire for a service implemented by Composite
> -----------------------------------------------------------------
>
>                 Key: TUSCANY-2189
>                 URL: https://issues.apache.org/jira/browse/TUSCANY-2189
>             Project: Tuscany
>          Issue Type: Bug
>          Components: Java SCA Core Runtime
>            Reporter: Scott Kurz
>             Fix For: Java-SCA-Next
>
>         Attachments: test.2189.jar
>
>
> Take something like this:
> <composite name="OuterComposite">
>     <component name="OuterCalculatorComponent">
>       <service name="OuterCalculatorService">
>         <binding.ws wsdlElement="...."/>
>       </service>
>       <implementation.composite name="calc:InnerComposite"/>
>     </component>
> </composite>
> <composite name="InnerComposite">
>     <service name="OuterCalculatorService" 
> promote="CalculatorComponent/CalculatorService"/>
>     <component name="CalculatorComponent">
>         <service name="CalculatorService"/>
>         <implementation.java class="calculator.CalculatorServiceImpl"/>
>     </component>
> </composite>
> ------------------------------------------
> I'm noticing that the wireTarget that ends up getting built for the wire from 
> the OuterCalculatorService service-side WS binding into the impl has a 
> wireTarget
> with a Composite impl.     This causes a problem when 
> RuntimeWireImpl.initInvocationChains()  calls addImplementationInterceptor(); 
> we need a non-composite impl (Java impl) at this point to set up the 
> interceptor on the chain.
> Might it be appropriate to do something like what's done in 
> CompositeWireBuilderImpl.connectComponentReferences(), where we drill down 
> recursively to unwrap the Composite impl services?
> I looked at the 'recursive' itest and didn't see anything besides 
> binding.sca... so maybe we don't think we've gotten to this yet.
> In the cases it does work, with binding.sca, we must somehow be connected to 
> the "inner"-layer wire from the component in the nested Composite.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to