[ 
http://issues.apache.org/jira/browse/TUSCANY-833?page=comments#action_12441862 
] 
            
Jim Marino commented on TUSCANY-833:
------------------------------------

Looking into this more, my preference would be not to fix this or do a hack for 
M2 since for POJOs, component type side files aren't really useful do to some 
spec issues. Even if we were to fix the problems outlined, the only things that 
can be specified in a side file are services, references and proprties. The 
spec has not yet defined how to represent specific Java concepts such as 
init/destroy and implementation scopes in the component type schema. This 
renders side files not very useful.  The only type of POJO implementation that 
could be specified with a side file is statelesss with no intitialization or 
destruction callbacks. In addition, we already support the algorithm for 
determining services, references and properties on an unannotated POJO, 
relegating the need to actuallly have to specify a side file to mostly corner 
cases involving legacy code.

Given the need to stabilize a brach, I think we should release note that we do 
not support component type side files for POJOs with M2.   

> ComponentType sidefiles do not work for Pojo Implementation
> -----------------------------------------------------------
>
>                 Key: TUSCANY-833
>                 URL: http://issues.apache.org/jira/browse/TUSCANY-833
>             Project: Tuscany
>          Issue Type: Bug
>          Components: Java SCA POJO Container
>    Affects Versions: Java-M2
>            Reporter: Venkatakrishnan
>         Assigned To: Jim Marino
>            Priority: Critical
>
> If you have a component type sidefile for a Pojo implementation we end up 
> with an exception.  The reason for this is that the JavaComponentTypeLoader 
> passes the PojoComponenType.class to the loader registry to be returned as a 
> result.  However what gets created is an instance of the base ComponentType 
> and then there is an attempt to narrrow this to a PojoComponentType which 
> results in an exception.
> A quick alternative in the interest of M2 fast approaching would be to take 
> the approach that the containers have to get over this problem which is for 
> the containers to get the base ComponentType and copy it over to the special 
> ones.

-- 
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

        

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

Reply via email to