Mailing-List: contact axis-dev-help@ws.apache.org; run by ezmlm
Precedence: bulk
Reply-To: axis-dev@ws.apache.org
Received-SPF: neutral (asf.osuosl.org: local policy)
Message-ID: <43CCF2B5.4040301@apache.org>
Date: Tue, 17 Jan 2006 13:35:49 +0000
From: Steve Loughran <stevel@apache.org>
User-Agent: Thunderbird 1.5 (X11/20051201)
MIME-Version: 1.0
To: axis-dev@ws.apache.org
Subject: Re: [Axis2] New SOAP Fault implementation Questions
References: <43BA17AD.1010804@opensource.lk>
In-Reply-To: <43BA17AD.1010804@opensource.lk>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Eran Chinthaka wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>  
> 
> Hi,
> 
> I think Steve started some time back on implementing faults for Axis2.
> And I think its not completed and integrated. Shall we remove that
> code from this release also, as its not being used ? I have few other
> questions on impl deatails as well.
> 
>    1. Why is the design completely stands away alone from OM. The impl
>       I did was an extension OM. IMO, if we create a separate object
>       model for faults, we will have two object models for faults. I
>       hope everyone agrees that we need to have the OM structure. So
>       if we can merge that structure with Steve has defined, we can
>       avoid duplication.

One of the problems with OM is its binding to SOAP1.1 or 1.2. you have 
to choose early on, what kind of fault you throw/catch. I need to 
generate faults without caring about its outbound transport. Relays may 
also want to catch a fault, fiddle with it and pass it on.

However, its hard to be soap version independent in Axis, because you 
cant add nested fault detail elements without knowing where they are 
going. So arguably, the design is not the correct one for Axis (although 
it is for any Xom-based system)

>    2. The AbstractFaultCode class maintains an instance of its
>       subclass FaultSubCode and some references to that. I don't think
>       its a good decision, unless you have a very good reason.

I am trying to remember.