W3C home > Mailing lists > Public > www-ws-desc@w3.org > July 2006

Re: Request 2 clarifications for RPC style

From: Roberto Chinnici <Roberto.Chinnici@Sun.COM>
Date: Thu, 06 Jul 2006 10:42:48 -0700
To: Lawrence Mandel <lmandel@ca.ibm.com>
Cc: www-ws-desc@w3.org
Message-id: <44AD4B98.8010608@sun.com>

Hi Lawrence,

I agree with you on #1.

On #2, local elements do have qualified names. If an 
elementForm="qualified" or
elementFormDefault="qualified" attribute appears in the appropriate 
place in the schema,
then the {target namespace} property of the Element Declaration 
component for a
local element will be present, its value being the target namespace for 
the schema itself.

Based on what the spec currently says, I don't think you can infer that 
two local elements
with the same unqualified names and defined in the input and output 
sequences of a single
RPC-style operation will necessarily have the same qualified names.

Additionally, since the use of elementFormDefault="qualified" is quite 
common and in
many ways desirable, I'd recommend against changing the WSDL 2.0 spec to 
require local
elements to be unqualified (i.e. for their {target namespace} property 
to be absent).

Thanks,
Roberto

Lawrence Mandel wrote:
>
> While creating tests for the RPC style I came across two assertions 
> that have ambiguous language even though I think I understand the 
> meaning and that I'd like to see clarified. (I think the 
> clarifications are simple editorial changes to remove any doubt as to 
> the meaning of the assertions.)
>
> 1. "The sequence MUST contain only local element children._^ _ 
> <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-adjuncts.html?content-type=text/html;%20charset=utf-8#RPCStyle-5014-summary> 
> " [1]
>
> I believe this assertion refers to both input element and output 
> element sequences. The other assertions for RPC style explicitly state 
> input or output. I'd like to see this assertion hardened as:
>
> "Both the input and output sequences MUST contain only local element 
> children."
>
> 2. "If elements with the same qualified name appear as children of 
> both the input and output elements, then they MUST both be declared 
> using the same named type._^ _ 
> <http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/ws/desc/wsdl20/wsdl20-adjuncts.html?content-type=text/html;%20charset=utf-8#RPCStyle-5018-summary> 
> " [2]
>
> In this case local elements do not have qualified names only local 
> names. This looks like a mistake to me. I'd like to see this corrected 
> to avoid any confusion. My suggested correction is:
>
> "If elements with the same local name appear...."
>
> I've attached a patch for wsdl20-adjunts.xml that contains these changes.
>
> [1] 
> http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl20/wsdl20-adjuncts.html?content-type=text/html;%20charset=utf-8#RPCStyle-5013-summary 
>
> [2] 
> http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl20/wsdl20-adjuncts.html?content-type=text/html;%20charset=utf-8#RPCStyle-5018-summary 
>
>
> Thanks,
>
> Lawrence Mandel
>
> Software Developer
> IBM Rational Software
> Phone: 905 - 413 - 3814 Fax: 905 - 413 - 4920
> lmandel@ca.ibm.com
>
Received on Thursday, 6 July 2006 17:43:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:41 GMT