RE: ACTION-64 - Clarify spec about topic replyToName

Greetings Peter, looks like a good starting position. Comments inline . . 
. 

Regards, Roland


public-soap-jms-request@w3.org wrote on 23/02/2009 23:55:44:

> [image removed] 
> 
> RE: ACTION-64 - Clarify spec about topic replyToName
> 
> Rokicki, Derek 
> 
> to:
> 
> Rokicki, Derek, public-soap-jms
> 
> 23/02/2009 23:58
> 
> Sent by:
> 
> public-soap-jms-request@w3.org
> 
> Group,
> As I mentioned in our February 9th meeting, it seems that we 
> overlooked the idea of a topicReplyToName in the SOAP over JMS 1.0 
> specification.
> According to section 4.3.1. of the URI Scheme, if a JMS topic is 
> required as a response destination, then a JMS URI can employ the 
> "topicReplyToName" parameter.  This parameter is not mentioned in 
> the SOAP over JMS 1.0 specification at all.  For consistency sake I 
> suggest we modify section 2.2.2 of the SOAP over JMS 1.0 
> specification.  Here is one possible set of changes:
> Add the following comments to the soapjms:ReplyToName definition:
> ·       If the variant is "queue" or "topic", the replyToName 
> parameter always refers to a name of a JMS queue.
> ·       Both the "replyToName" and "topicReplyToName" parameters 
> MUST not be specified at the same time.
> Add a new definition:
> [Definition: soapjms:topicReplyToName] (xsd:string) 
> ·       Specifies the name of the topic destination to which a 
> response message should be sent.
> ·       If the variant is "jndi", then this value is ignored.
> Question, should we ignore the value or should we treat it like we 
> do the replyToName?

topicReplyToName is not appropriate with "jndi" variant so I propose that 
it should simply be ignored.

> ·       Both the "replyToName" and "topicReplyToName" parameters 
> MUST not be specified at the same time.
> Question, what if they are?  Do we ignore one or does this cause a 
fault?

Suggest that MUST NOT becomes SHOULD NOT and define which will be used in 
a set of enumerated circumstances.

> Question, what if replyToName is in one source and topicReplyToName 
> is in another?

one of the enumerated circumstances.

> ·       optional in URI, optional in WSDL, optional in environment
> ·       if specified, this is used to derive the value to be used in
> the JMS header JMSReplyTo
> If there is time, I would like to discuss this further in tomorrow?s 
meeting.
> Thanks,
> Derek





Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU