Re: Another proposal for async extensions

proton@netspace.net.au wrote:

>What if I want replyTo to be specified, but faultTo to be anonymous? Must I 
>specify "both"? Even if I never want anonymous replyTo, and never want non-
>anonymous faultTo?
>  
>
Yes.

>I know it's an irritating case, but I'd like to bring it up while the 
>discussion is hot :-)
>
>Tony
>
>Quoting Anish Karmarkar <Anish.Karmarkar@oracle.com>:
>
>  
>
>>Yalcinalp, Umit wrote:
>>    
>>
>>>
>>>      
>>>
>>>>-----Original Message-----
>>>>From: Marc Hadley [mailto:Marc.Hadley@Sun.COM]
>>>>Sent: Thursday, Jun 16, 2005 12:17 PM
>>>>To: Yalcinalp, Umit
>>>>Cc: public-ws-async-tf@w3.org
>>>>Subject: Re: Another proposal for async extensions
>>>>
>>>>On Jun 16, 2005, at 3:06 PM, Yalcinalp, Umit wrote:
>>>>
>>>>        
>>>>
>>>>>>2.1 Extend the existing wsaw:UsingAddressing Element
>>>>>>
>>>>>>Add an attribute 'asyncOnly' with a default value of 'false'. When
>>>>>>'true' the endpoint only supports async interactions.
>>>>>>
>>>>>>
>>>>>>            
>>>>>>
>>>>>How about supporting async, noAsync, or both?
>>>>>
>>>>>We would like an endpoint to support all three cases. Would you
>>>>>consider
>>>>>that? This is in essence the guts of our original proposal...
>>>>>
>>>>>          
>>>>>
>>>>You mean something like:
>>>>
>>>><wsaw:UsingAddressing responseChannel="anonymous|specified|both">
>>>>  <wsaw:ResponseBinding>...</wsaw:responseBinding>
>>>></wsaw:UsingAddressing>
>>>>
>>>>where responseChannel="anonymous" means [reply endpoint] and [fault
>>>>endpoint] can only be anonymous, responseChannel="specified" means
>>>>[reply endpoint] and [fault endpoint] can not be anonymous, and
>>>>responseChannel="both" means [reply endpoint] and [fault endpoint]
>>>>can either be anonymous or something else (as further constrained by
>>>>the child wsaw:ResponseBinding elements).
>>>>
>>>>        
>>>>
>>>Yes. This was pretty much what we were going to propose collectively
>>>which I have not got around writing...
>>>
>>>      
>>>
>>If we proceed in this direction, which I do like, it would be good to
>>have the same (or very similar) WSDL decorations for both SOAP 1.1 and
>>SOAP 1.2.
>>
>>-Anish
>>--
>>
>>    
>>
>>>>Works for me.
>>>>
>>>>Marc.
>>>>
>>>>
>>>>        
>>>>
>>>>>>Add a wsaw:ResponseBinding child element with cardinality
>>>>>>[0..unbounded]. The value of each of these is a binding
>>>>>>identification URI that specifies that the given endpoint can
>>>>>>support
>>>>>>[reply endpoint] and [fault endpoint] destinations using the
>>>>>>appropriate binding. If wsaw:UsingAddressing/@asyncOnly='true' then
>>>>>>there must be at least one
>>>>>>            
>>>>>>
>>>>wsaw:UsingAddressing/wsaw:ResponseBinding
>>>>
>>>>        
>>>>
>>>>>>element.
>>>>>>
>>>>>>If there are zero
>>>>>>            
>>>>>>
>>>>wsaw:UsingAddressing/wsaw:ResponseBinding elements
>>>>
>>>>        
>>>>
>>>>>>then the only [destination] supported for [reply endpoint]
>>>>>>and [fault
>>>>>>endpoint] is the anonymous URI.
>>>>>>
>>>>>>[1] http://www.w3.org/TR/2003/REC-soap12-part2-20030624/#soapinhttp
>>>>>>[2] http://dev.w3.org/cvsweb/~checkout~/2004/ws/addressing/ws-addr-
>>>>>>soap.html?content-type=text/html;%20charset=utf-8#s12feature
>>>>>>[3] http://dev.w3.org/cvsweb/~checkout~/2004/ws/addressing/ws-addr-
>>>>>>soap.html?content-type=text/html;%20charset=utf-8#s12module
>>>>>>[4] http://www.w3.org/TR/2003/REC-soap12-part2-20030624/
>>>>>>#tabreqstatereqtrans
>>>>>>[5] http://www.w3.org/TR/2003/REC-soap12-part2-20030624/
>>>>>>#tabresstaterecheads
>>>>>>
>>>>>>---
>>>>>>Marc Hadley <marc.hadley at sun.com>
>>>>>>Business Alliances, CTO Office, Sun Microsystems.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>            
>>>>>>
>>>>>          
>>>>>
>>>>---
>>>>Marc Hadley <marc.hadley at sun.com>
>>>>Business Alliances, CTO Office, Sun Microsystems.
>>>>
>>>>
>>>>
>>>>        
>>>>
>>>      
>>>
>>    
>>
>
>
>
>
>
>------------------------------------------------------------
>This email was sent from Netspace Webmail: http://www.netspace.net.au
>
>
>
>  
>

Received on Thursday, 23 June 2005 14:06:02 UTC