You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cxf.apache.org by Benson Margulies <bi...@gmail.com> on 2008/01/22 03:30:48 UTC
Fun with mtom
Good news: I've wired @MTOM all the way down to the
JAXBAttachmentMarshaller.
Bad news: I introduced a bit of a dependency from the JAXBDataBinding on
JAXWS, by actually using the MTOMFeature class to carry the facts
downstream.
If this really bothers someone, it can clearly be replaced by some class
of ours. As usual, I'm not a fan of the 'big bag of properties', so I'd
make AbstractDataBinding responsible for hanging onto the boolean and
integer that are at issue here.
Re: Fun with mtom
Posted by Benson Margulies <bi...@gmail.com>.
On Tue, 2008-01-22 at 13:04 +0000, Sergey Beryozkin wrote:
> Hi Benson
>
> Many thanks. The reason I asked the question is that JAX-RS and JAX-WS, both relying on JAXB in some way, are often seen as opposing technologies, so going forward, it might be worth it to provide a number of bundles for different types of users, say, all CXF, JAXWS-only, JAX-RS only...
That all makes sense, you're welcome. I'm not thrilled with the
modularity, but at least I avoided the gross dependency.
Hypothetically, any data binding can do MTOM, so there needs to be some
cross-data-binding configuration. Any front end can control MTOM.
Re: Fun with mtom
Posted by Sergey Beryozkin <se...@iona.com>.
Hi Benson
Many thanks. The reason I asked the question is that JAX-RS and JAX-WS, both relying on JAXB in some way, are often seen as opposing technologies, so going forward, it might be worth it to provide a number of bundles for different types of users, say, all CXF, JAXWS-only, JAX-RS only...
Cheers, Sergey
----- Original Message -----
From: "Benson Margulies" <bi...@gmail.com>
To: <cx...@incubator.apache.org>
Sent: Tuesday, January 22, 2008 12:55 PM
Subject: Re: Fun with mtom
>
> On Tue, 2008-01-22 at 10:30 +0000, Sergey Beryozkin wrote:
>> > Bad news: I introduced a bit of a dependency from the JAXBDataBinding on
>> > JAXWS, by actually using the MTOMFeature class to carry the facts
>> > downstream.
>>
>> So, what if we were to repackage a CXF a bit such that say people interested in JAX-RS only could get a bundle
>> which contains only JAX-RS and JAXB. Would your change prevent us from doing so ?
>>
>> Thanks, Sergey
>
> It would have if I'd stuck with it. I reworked it to remove the
> dependency before I checked it in.
>
----------------------------
IONA Technologies PLC (registered in Ireland)
Registered Number: 171387
Registered Address: The IONA Building, Shelbourne Road, Dublin 4, Ireland
Re: Fun with mtom
Posted by Benson Margulies <bi...@gmail.com>.
On Tue, 2008-01-22 at 10:30 +0000, Sergey Beryozkin wrote:
> > Bad news: I introduced a bit of a dependency from the JAXBDataBinding on
> > JAXWS, by actually using the MTOMFeature class to carry the facts
> > downstream.
>
> So, what if we were to repackage a CXF a bit such that say people interested in JAX-RS only could get a bundle
> which contains only JAX-RS and JAXB. Would your change prevent us from doing so ?
>
> Thanks, Sergey
It would have if I'd stuck with it. I reworked it to remove the
dependency before I checked it in.
Re: Fun with mtom
Posted by Sergey Beryozkin <se...@iona.com>.
> Bad news: I introduced a bit of a dependency from the JAXBDataBinding on
> JAXWS, by actually using the MTOMFeature class to carry the facts
> downstream.
So, what if we were to repackage a CXF a bit such that say people interested in JAX-RS only could get a bundle
which contains only JAX-RS and JAXB. Would your change prevent us from doing so ?
Thanks, Sergey
> Good news: I've wired @MTOM all the way down to the
> JAXBAttachmentMarshaller.
>
> Bad news: I introduced a bit of a dependency from the JAXBDataBinding on
> JAXWS, by actually using the MTOMFeature class to carry the facts
> downstream.
>
> If this really bothers someone, it can clearly be replaced by some class
> of ours. As usual, I'm not a fan of the 'big bag of properties', so I'd
> make AbstractDataBinding responsible for hanging onto the boolean and
> integer that are at issue here.
>
----------------------------
IONA Technologies PLC (registered in Ireland)
Registered Number: 171387
Registered Address: The IONA Building, Shelbourne Road, Dublin 4, Ireland