W3C home > Mailing lists > Public > public-ws-addressing-tests@w3.org > December 2005

RE: WS- Addressing "Hello World"

From: <paul.downey@bt.com>
Date: Thu, 8 Dec 2005 13:54:02 -0000
Message-ID: <2A7793353757DB4392DF4DFBBC952255E907FD@I2KM11-UKBR.domain1.systemhost.net>
To: <Arun.Gupta@Sun.COM>
Cc: <Mike.Vernal@microsoft.com>, <public-ws-addressing-tests@w3.org>

Hi Arun!

> Since WSDL Binding is not in scope for this round of CR testing, I'm 
> assuming it should not matter if either WSDL 1.1 or WSDL 2.0 
> is used to generate the messages. Is that correct ?

They're just descriptions of the messages exchanged intended
for your convenience. You can use one of the WSDLs, both or ignore
them completely, all that matters is that the *messages* exchanged 
satisfy the XPath assertions in the testcases.

> Another related question, the targetNamespace for the WSDL is 
> "http://example.com/wsaTestService". A targetNamespace such as 
> "urn:example.com:wsaTestService" (using ":" separator instead of "/") 
> would be tested as part of WSDL CR testing. Is that correct 
> or we need to worry about that now as well ?

You can use a WSDL with a different namespace, portTypes and even 
different operation names and pass all of the tests - it's all just 
abstract stuff local to the description and should have no impact 
on the actual messages exchanged.

> Is it fair to assume that the testcase numbers assigned will 
> not change going forward ? Thus new test cases will be added without 
> disturbing the existing numbering scheme.

That's the plan. Also if we remove a testcase, we should ensure
we don't reuse that number for a different test to avoid confusion.

Received on Thursday, 8 December 2005 13:54:28 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:29:01 UTC