Re: SOAP 1.1 One-way HTTP Binding doc

David Orchard wrote:

>Soap-Request-optional-soap-response vs
>soap-request-optional-protocol-response?  
>SOAP Request-optional-response vs SOAP
>request-optional-protocol-response?
>  
>I don't see why we need to distinguish between a SOAP response with 200
>vs with 202.  
>
>Another way of looking at this is there are 2 simultaneous and
>orthogonal changes going on: 
>1) adding 202 as a valid status code for the soap 1.1 http binding for
>request-response.
>2) adding a one-way that allows for no soap response that uses 202 as
>the status code.
>  
>
These don't look orthogonal to me.  Orthogonal would be allowing for
empty body independently of allowing for 202 (all four options allowed),
having empty/non-empty mean one thing and 200/202 mean something
completely separate.

We've been pretty clear for a while that empty 202 means "ack".  I'm
hearing that non-empty 202 is meant for things like WS-RX acks, but I'm
not sure this is nailed down.  There seems to be some possibility that a
202 with a SOAP envelope could also be a real response.

If 202 can be a real response, then one would have to use something
besides 202 to figure out what's really going on (e.g., whether the
message consists only of WS-RX headers and similar).  In this case 202
isn't really carrying any information and why bother allowing for it? 
On the other hand, if 202 means something in particular, then what
exactly does it mean?

As far as I can tell, the value in non-empty 202 is telling the SOAP
stack "Hey, this is just infrastructure stuff.  Don't pass it along to
the application."  We can't say that here, but we could (probably) say
it elsewhere.

>Cheers,
>Dave
>
>  
>
>>-----Original Message-----
>>From: Anish Karmarkar [mailto:Anish.Karmarkar@oracle.com]
>>Sent: Monday, January 30, 2006 12:59 PM
>>To: David Orchard
>>Cc: Mark Baker; Christopher B Ferris; WS-Addressing
>>Subject: Re: SOAP 1.1 One-way HTTP Binding doc
>>
>>David Orchard wrote:
>>    
>>
>>>There is a lot of level mixing going on.  I don't think that an
>>>      
>>>
>optional
>  
>
>>>SOAP response should be part of a "one-way" binding.  Sure, it might
>>>      
>>>
>be
>  
>
>>>a WSDL one-way, but we are talking about SOAP not WSDL.  SOAP
>>>definitions shouldn't be coupled up the stack to WSDL, and I'm
>>>      
>>>
>against
>  
>
>>>defining SOAP things in the context of WSDL.  SOAP shouldn't know
>>>anything about the stuff that is describing it, beit WSDL, Policy,
>>>Semantic Web, foo...
>>>
>>>      
>>>
>>I agree, that (WSDL-independence) is a fine goal to have.
>>
>>But, a SOAP response in a SOAP request-response exchange (status code
>>200) is different from a SOAP response in a SOAP
>>request-optional-response exchange (status code 202) -- it should not
>>    
>>
>be
>  
>
>>construed that a SOAP request-optional-response exchange for the case
>>when the SOAP envelope is sent back in a 202 HTTP response is an
>>'instance' of the SOAP request-response exchange. Perhaps there is a
>>better way to name the "MEP" so as to disambiguate this
>>(request-optional-ack or something like that).
>>
>>-Anish
>>--
>>
>>    
>>
>>>Again, it's an optional SOAP envelope, so in the context of a SOAP
>>>binding it should be called optional response.  In the case of a
>>>      
>>>
>WSDL
>  
>
>>>one-way, I'd see that the complete description, including things
>>>      
>>>
>like
>  
>
>>>Policy assertions of RM-level acks and WSDL descriptions, is
>>>      
>>>
>required to
>  
>
>>>accurately describe whether a SOAP envelope is allowed as a response
>>>      
>>>
>or
>  
>
>>>not.
>>>
>>>Cheers,
>>>Dave
>>>
>>>
>>>      
>>>
>>>>-----Original Message-----
>>>>From: Anish Karmarkar [mailto:Anish.Karmarkar@oracle.com]
>>>>Sent: Monday, January 30, 2006 12:26 PM
>>>>To: David Orchard
>>>>Cc: Mark Baker; Christopher B Ferris; WS-Addressing
>>>>Subject: Re: SOAP 1.1 One-way HTTP Binding doc
>>>>
>>>>Thinking more on this, isn't this still a one-way?
>>>>I.e., a SOAP envelope can come back on the HTTP 202 response without
>>>>making it a request-response.
>>>>
>>>>202 is intentionally non-committal. It says 'Accepted'. A RM-level
>>>>        
>>>>
>ack
>  
>
>>>>does not mean that the SOAP envelope is a 'response' to the
>>>>        
>>>>
>'request'
>  
>
>>>in
>>>
>>>      
>>>
>>>>the HTTP request.
>>>>
>>>>I think it is fine to call it one-way (as you did in your previous
>>>>formulation). This is important, as there aren't any SOAP MEPs in
>>>>        
>>>>
>SOAP
>  
>
>>>>1.1 so everything is in the context of a WSDL operation. In the case
>>>>        
>>>>
>>>of
>>>
>>>      
>>>
>>>>status code 202, there isn't a WSDL level response as it is a WSDL
>>>>one-way operation.
>>>>
>>>>-Anish
>>>>--
>>>>
>>>>David Orchard wrote:
>>>>
>>>>        
>>>>
>>>>>Sounds like it's a request-optional response HTTP binding that
>>>>>          
>>>>>
>y'all
>  
>
>>>are
>>>
>>>      
>>>
>>>>>looking for.
>>>>>
>>>>>Dave
>>>>>
>>>>>
>>>>>
>>>>>          
>>>>>
>>>>>>-----Original Message-----
>>>>>>From: mbaker@gmail.com [mailto:mbaker@gmail.com] On Behalf Of Mark
>>>>>>            
>>>>>>
>>>>>Baker
>>>>>
>>>>>
>>>>>          
>>>>>
>>>>>>Sent: Friday, January 20, 2006 11:41 AM
>>>>>>To: David Orchard
>>>>>>Cc: Christopher B Ferris; WS-Addressing
>>>>>>Subject: Re: SOAP 1.1 One-way HTTP Binding doc
>>>>>>
>>>>>>On 1/20/06, David Orchard <dorchard@bea.com> wrote:
>>>>>>
>>>>>>
>>>>>>            
>>>>>>
>>>>>>>So y'all are looking for a binding that says a 202 is allowed and
>>>>>>>              
>>>>>>>
>>>if
>>>
>>>      
>>>
>>>>>so,
>>>>>
>>>>>
>>>>>          
>>>>>
>>>>>>>the response may or may not contain a SOAP envelope.  It's the
>>>>>>>preclusion of the soap envelope that's the problem?
>>>>>>>              
>>>>>>>
>>>>>>From my POV, yep!
>>>>>
>>>>>          
>>>>>
>>>>>>Mark.
>>>>>>            
>>>>>>
>>>>>
>>>>>          
>>>>>
>
>
>  
>