W3C home > Mailing lists > Public > www-ws-desc@w3.org > January 2007

RE: Dispatching GuaranteedFault (MessageTest-1G)

From: Jonathan Marsh <jonathan@wso2.com>
Date: Wed, 17 Jan 2007 11:21:39 +0530
To: "'www-ws-desc'" <www-ws-desc@w3.org>
Message-ID: <003c01c739fb$915cefb0$d901640a@DELLICIOUS>
I've done this, as follows:


1)      Removed GenericSOAPBinding from MessageTest-1G, including the
endpoint referring to it.

2)      Created MessageTest-6G that only has the GenericSOAPBinding,
including removing endpoints associated with other bindings.

3)      Changed the content models of Send and GuaranteedFault to elements
sendString and sendInt, with xs:string and xs:int types repectively.

4)      Changed the name of the Service, changed some comments


That's about it.  Hopefully the bogus red resulting from this surgery will
evaporate quickly.


Jonathan Marsh -  <http://www.wso2.com> http://www.wso2.com -
<http://auburnmarshes.spaces.live.com> http://auburnmarshes.spaces.live.com



From: www-ws-desc-request@w3.org [mailto:www-ws-desc-request@w3.org] On
Behalf Of Jonathan Marsh
Sent: Wednesday, January 10, 2007 11:27 AM
To: 'www-ws-desc'
Subject: Dispatching GuaranteedFault (MessageTest-1G)


There is an unfortunate interaction in MessageTest-1G, with the
GuaranteedFault and Send operations when bound to the GenericSOAPBinding.


These operations test the #any and #none content models.  Since the payload
can be any XML, or no XML, and there are no per-operation details specified,
these messages can't be property dispatched by Axis2.  The mechanisms we
typically use (wsa:Action, {soap action}, or unique GEDs, {http location})
aren't available.


The problem arises in that the test tries to simultaneously stress test
constructs like #any, #none, and generic bindings.  Testing these features
in isolation prevents the unfortunate interaction.


To solve this, I propose we untease these parts of the test, in the
following way:


1)  Remove the GenericSOAPBinding from MessageTest-1G.

2)  Create a new test GenericBinding-1G which binds the interesting types of

a.   in-out MEP

b.   in-out MEP with returned fault

c.   in-only MEP

d.   robust-in-only MEP

e.   robust-in-only MEP with returned fault.


I'm happy to do this work if the implementers agree it's useful.  Let's
discuss on Thursday.


Jonathan Marsh -  <http://www.wso2.com> http://www.wso2.com -
<http://auburnmarshes.spaces.live.com> http://auburnmarshes.spaces.live.com


Received on Wednesday, 17 January 2007 05:51:55 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 23:07:05 UTC