I went ahead and added a utility JAXRSDataBinding which can be configured
with a custom JAX-RS provider reference.

The custom provider must implement MessageBodyWriter but is not required to
implement MessageBodyReader (in case service methods have no input
parameters).

Give it a try please.

cheers, Sergey

Please see [1] and [2]
[1]
https://cwiki.apache.org/confluence/display/CXF20DOC/JAX-RS+Data+Bindings#JAX-RSDataBindings-JAXRSDataBinding
[2] http://svn.apache.org/viewvc?rev=1054645&view=rev




On Tue, Dec 28, 2010 at 11:06 PM, Sergey Beryozkin <[email protected]>wrote:

> Hi
>
> On Tue, Dec 28, 2010 at 7:31 PM, ntchoate <[email protected]> wrote:
>
>>
>>
>> Sergey Beryozkin-5 wrote:
>> >
>> >
>> > Does it help ? Or do you need to use your own custom MessageBodyWriter ?
>> > Ex,
>> > JAXBElementProvider with some configuration properties which are not
>> > available with the JAXBDataBinding ?
>> >
>> > cheers, Sergey
>> >
>>
>> I think you confirmed what I already thought I would have to resort to.
>>  I'm
>> not a big fan of JAXB, and thought it was much simpler to write my own
>> small
>> XML marshallers utilizing the STaX API, since each marshaller was only
>> about
>> 4 lines of code and I only needed a few.  In addition, by using the STaX
>> API, I could leverage jettison to produce JSON output using my existing
>> XML
>> marshalling classes.  I was really just looking for a way to keep my
>> XML/JSON format separate from my object model.
>>
>> If I were to boil my question down to a more succinct version, will CXF in
>> the future support a similar serialization mechanism between jax-rs and
>> jax-ws (MessageBodyWriter/Reader vs DataBinding), or are the specs simply
>> incompatible to support such a feature?
>>
>
> I think it can work. We can probably consider introducing CXF
> DelegatingDataBinding that CXF JAXWS can work with directly which would
> delegate to arbitrary CXF DataBindings.Next, CXF JAX-RS would provide an
> implementation of CXF DataBinding which wouild simply delegate to
> MessageBodyReader/Writer. We need to think a bit more about it...
>
> cheers, Sergey
>
>
>
>
>> --
>>
>> View this message in context:
>> http://cxf.547215.n5.nabble.com/MessageBodyWriters-and-SOAP-tp3319517p3320617.html
>> Sent from the cxf-user mailing list archive at Nabble.com.
>>
>
>

Reply via email to