RE: Patch for testcase wsoap:header with required=false

I think the separate SOAPHeader-1G testcase is better, as implementations
can either insert the header or not leading to various results, one of which
doesn't really test that the header declaration is ignored, and the other
one, well, simply ignores it.  I think we already decided that message
testing on this wouldn't make much sense - probably over dinner back in
Rennes?

 

I'll check in the new testcase and regenerate the coverage report results.

 

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 John Kaputin (gmail)
Sent: Thursday, February 01, 2007 3:11 PM
To: www-ws-desc
Cc: woden-dev@ws.apache.org
Subject: Patch for testcase wsoap:header with required=false

 

I have an action item in the Minutes of the weekly call 25 Jan to create a
test case for the last outstanding test on the Interchange Test Coverage
Report.  This is to test wsoap:header with required=false.   

Two candidate test cases are attached as patch files: 

1) If you want to include required=false in the message interop testing,
then use MessageTest-5G.patch.txt. This patch file adds an operation to this
WSDL with required=false in the wsoap:header element in each binding. I
could create a new separate testcase, MessageTest-7G, if you prefer.

2) If you just want to check that implementations can parse required=false
in 'good' wsdl (but don't need to use it in the message interop testing),
then use SOAPHeader-1G.patch.txt. This is a cut down version of
MessageTest-5G with just the operation that contains the wsoap:header with
required=false.

regards,
John Kaputin.