Mailing-List: contact dev-help@cxf.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cxf.apache.org
Received-SPF: pass (athena.apache.org: domain of elakito@googlemail.com
 designates 209.85.160.169 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=googlemail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type:content-transfer-encoding;
        b=YRyjR8rF4inqQ2d/37aSbSDPxe/gfG0oIhPTDvT388yASLCDi+u3Bqnwgy9ytdYXl/
         T2FVciax2n32X+NlpFvK+LuoWsEBRF5H9mfukXt6E56X7Wqe5bJ9A5hSdLgAfx0MQcWE
         rabXsrFOrUBd4uq/ZxbtJJcmEVfM1iJ6YRlqI=
MIME-Version: 1.0
In-Reply-To: <4D915C5D.7070901@sosnoski.com>
References: <4D8B319D.5050207@sosnoski.com>
	<201103240949.34353.dkulp@apache.org>
	<4D8C7443.2040907@sosnoski.com>
	<BANLkTindnsURyo_UyN2a1GbLkOv+LGdJ=w@mail.gmail.com>
	<4D8D4A40.7040807@sosnoski.com>
	<4D8D510D.6060106@sosnoski.com>
	<BANLkTimXEOcbt2zH0u8oA1_8CpU2wVwZcg@mail.gmail.com>
	<BANLkTi=oLy6F-buHV-4Z+vFN0W93DHEjGg@mail.gmail.com>
	<4D906831.6020706@sosnoski.com>
	<BANLkTincD7p9sKOboeJ2VOvhpxwCKMF7Ug@mail.gmail.com>
	<4D915C5D.7070901@sosnoski.com>
Date: Tue, 29 Mar 2011 14:02:33 +0200
Message-ID: <AANLkTikpzXRYuqnpdmO0opM=kj_BQsew17oHSz9G2YM5@mail.gmail.com>
Subject: Re: Adding value to WS-RM feature
From: Aki Yoshida <elakito@googlemail.com>
To: dev@cxf.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hi Dennis,
sounds good.

Regarding my remark on this usage (this combination of using wsrm 1.0
with wsa 2005/08), I was talking about this particular combination
itself. I think most people would go for the standard wsrm 1.0 (wsrm
1.0 with wsa 2004/08) or the standard wsrm 1.1 (wsrm 1.1 with wsa
2005/08). In other words, the only valid use case for this particular
combination would be for communicating with some partial wsrm 1.0 or
1.1 implementation. So, from the interoperability robustness point of
view, this combination seems useful, but otherwise, it has littlel
use.

Regards, aki

2011/3/29 Dennis Sosnoski <dms@sosnoski.com>:
> Hi Aki,
>
> On 03/29/2011 04:19 AM, Aki Yoshida wrote:
>>
>> ...
>> But in any case, if we don't go for this modified class approach, we
>> might take option 2.
>> I think this option isn't bad, considering that we need to duplicate
>> only CreateSequenceType and AcceptType. And these classes are probably
>> only needed until the WSRM 1.1 is implemented.
>
> I'm not sure why you'd say these are only needed until WS-RM 1.1 are
> implemented. My intention is not to replace WS-RM 1.0 support with 1.1/1.=
2
> (which would break compatibility with older versions of CXF, and with
> services configured to use WS-RM 1.0), but to add support for 1.1/1.2.
>
> But I think you're right that option 2 (generating separate CreateSequenc=
e
> and Accept elements - actually CreateSequence and CreateSequenceResponse
> elements, since Accept is embedded in the latter - using the WS-Addressin=
g
> 2005 namespace) is the right way to go, and I'll change my code to match.
> The deciding factor for me is that this same approach of converting data
> structure versions is going to be needed to support 1.1/1.2, so I may as
> well also use it for this variation of 1.0.
>
> Thanks,
>
> =A0- Dennis
>
>