1) I have a set of slides describing the RESTfult services using Tuscany SCA. I 
can share it.

2) The other perspective is that SCA doesn't want to reinvent all the wheels. 
IMO, using JAX-WS, JAX-RS annotations to provide the mapping in java for the 
certain metadata is beneficial to SCA.

Thanks,
Raymond
________________________________________________________________ 
Raymond Feng
[email protected]
Apache Tuscany PMC member and committer: tuscany.apache.org
Co-author of Tuscany SCA In Action book: www.tuscanyinaction.com
Personal Web Site: www.enjoyjava.com
________________________________________________________________

On Aug 27, 2010, at 10:44 AM, ant elder wrote:

> On Fri, Aug 27, 2010 at 6:40 PM, Luciano Resende <[email protected]> wrote:
>> On Fri, Aug 27, 2010 at 10:37 AM, ant elder <[email protected]> wrote:
>>> On Fri, Aug 27, 2010 at 6:13 PM, Luciano Resende <[email protected]> 
>>> wrote:
>>> 
>>>> 
>>>> I'd try:
>>>> 
>>>>           <tuscany:binding.rest name="rest">
>>>>                <tuscany:operationSelector.rpc />
>>>>                <tuscany:response>
>>>>                   <tuscany:wireFormat.json />
>>>>                </tuscany:response>
>>>>           </tuscany:binding.rest>
>>>> 
>>>> This would then be accessible via URL
>>>> 
>>>> <base service URI> ?method=<operation name>&parm1=<value>&parm2=<value>
>>>> 
>>> 
>>> I've tried that and still get the IllegalArgumentException: wrong
>>> number of arguments. Looking in the code thats caused by the loop at
>>> line 113 of RPCOperationSelectorInterceptor which is only adding
>>> things to the message if the operation parameters have the jaxrs
>>> QueryParam annotation, which this service doesn't use. Whats the
>>> intention, should the binding work with arbitrary services or only for
>>> for jaxrs classes?
>>> 
>>>   ...ant
>>> 
>> 
>> Another option is to add support for declaratively add the mapping,
>> which is in my todo list, but not yet implemented. But, stepping back,
>> why not use the JAX-RS annotations ?
>> 
> 
> Because one of the big points of SCA is that the service
> implementations are protocol neutral?
> 
> Can you say a bit about what you're thinking of for "support for
> declaratively add the mapping"?
> 
>    ...ant

Reply via email to