cxf-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sergey Beryozkin <>
Subject How to consume JMS services
Date Mon, 28 Feb 2011 11:10:03 GMT

I'm starting a new thread...

Comments inline.

On Fri, Feb 25, 2011 at 5:04 PM, Bill Burke <> wrote:
> On 2/25/11 10:51 AM, Sergey Beryozkin wrote:
>> In your opinion, why would such (Java) users prefer an HTTP centric
>> interface for consuming messages backed up by JMS stores, when they
>> just can do plain Java JMS ?
>> What do you think ?
> (Remember you asked what I think so....)
> I don't think people should use CXF nor Resteasy nor SOAP if their
> environment is all Java.  As you said, use native Java JMS protocols.

I was not advocating the use of the native Java JMS only. If CXF
JAX-RS proxies could do it then I'd advocate using the proxies. At the
moment they can not do it. Generally speaking we would like CXF be
used everywhere, on the client and on the server side because it will
let us work on providing the additional QOS stuff via client-side
interceptors, transformations, policies, and the auto-discovery. We
don't mislead the users here. Users are clever and can always go their
own way.

> The benefit of using a Messaging RESTful interface is for, IMO, to provide
> lightweight inter-language/platform support.  If you check out the HornetQ
> rest examples, specifically Python ones, you'll see they use the built in
> http client that comes with the language and nothing else.

+1 - this is a good reason for creating such an interface. However
we're not interested (I'm saying 'we' here, thus take it with a grain
of salt please) in encouraging CXF users to stop using CXF and start
using a Python library for consuming JMS services. What is the point
here for CXF ?

However, having a CXF JAX-RS client consuming JMS-messages using a
'JMS-centric' language can be of interest. I hope HornetQ users will
ask us about it one day :-)

> Through content-negotiation and JAX-RS providers we also support
> transformation of JMS client sent Java objects into the representation the
> REST client desires.  And vice versa, unmarshalling of representations to
> Java objects using JAX-RS.

This is neat.

> As a side note, I'll eventually be pulling out the API into a specification.
>  Probably submitting it as an Internet Draft at IETF so we can leverage a
> neutral and respected entity's specification process.  That way, anybody
> could implement and support it.

Sounds promising. The thing is we can't do anything about it at the
CXF level. We can only write a demo showing CXF JAX-RS consumers
working with this interface.

>> If people using HornetQ as a messaging solution could plugin CXF
>> JAX-RS to back up your interface then I'd be the first one who would
>> blog about it and promote it. On CXF lists we discuss CXF-based issues
>> or solutions.
> HornetQ REST could probably be ported quite easily to CXF as most of it is
> pure JAX-RS (minus config and the client framework).  Honestly, from the
> client perspective it has nothing to do with CXF or Resteasy.  The client is
> free and encouraged to use any HTTP client framework in any language they
> desire.

The client perspective is a different issue. As I said, I don't see
the point in us advocating the use of non-CXF client frameworks
because at the CXF level it does not make sense. CXF is not a
container, CXF is a web services framework and thus its success
depends on how widely it is being used. The use of the CXF JAX-RS on
the HornetQ server side is the one which would motivate me and again
hope HornetQ users will ask us about it one day :-).

> I think its fair to suggest to somebody asking about JMS integration
> solutions other than CXF.  We encourage such posts on resteasy mail list.

Sure - unless we reckon we can provide something meaningful at the CXF level ?

>  For example, I always defer to Jersey regardings WADL as I have zero
> interest in implementing or supporting it.

ok. FYI, CXF supports WADL too :-) and plans to expand on this support

thanks, Sergey

P.S. Bill - you are welcome to contribute and challenge us on the dev
list. Sorry if the previous thread caused you some grief :-). I
believe no-one meant anything more than just a protective remark. It
is obvious now it turned out to be a highly controversial one but hope
such a seasoned professional as you are can accept the attempted
clarifications and apologies which followed :-)

> --
> Bill Burke
> JBoss, a division of Red Hat

View raw message