- From: Doug Davis <dug@us.ibm.com>
- Date: Wed, 9 Nov 2005 17:08:19 -0500
- To: www-ws@w3.org
- Message-ID: <OF583BDDBA.4F6A2921-ON852570B4.0078E7B6-852570B4.00797496@us.ibm.com>
Diego and I chatted briefly off-line and the WSA spec allows for multiple wsa:RelatesTo headers w/o any restrictions on the contents of each (e.g. each can have a RelationshipType of wsa:Reply). One option would be to define a new value for the RelationshipType attribute of wsa:RelatesTo (e.g. wsp:Reply) so that they can be differentiated. However, the response message is in fact also a wsa:Reply so I believe the current version of the ws-polling spec is compliant with WSA but this issues is something that should probably be explored in the future. thanks -Doug > From: www-ws-request@w3.org [mailto:www-ws-request@w3.org] On Behalf Of > Diego Gonzalez > Sent: Wednesday, November 09, 2005 9:21 AM > To: www-ws@w3.org > Subject: ws-polling comment > Hi, > > I've been reading the spec and I found something quirky in the > section 3.2 with respect to wsa:RelatesTo usage. The sample response > message described in the document contains two wsa:RelatesTo attributes > > <s:Envelope ...> > <s:Header ...> > <wsa:Action>...</wsa:Action> > <wsa:MessageID>xs:anyURI</wsa:MessageID> > <wsa:RelatesTo>xs:anyURI</wsa:RelatesTo> ? > <wsa:RelatesTo>messageID from GetMessage</wsa:RelatesTo> > <wsa:To>...</wsa:To> > ... > </s:Header> > <s:Body ...> > ... > </s:Body> > </s:Envelope> > > The first one is OPTIONAL and the second MUST be present. I think > there is very confusing to distinguish between them, since multiple > WS stacks use the RelatesTo element to match the response message > for a pending response, having two will force the stack to look > twice for the message. > > I think that it will be very valuable to add a RelationshipType > attribute to the RelatesTo element that represents the message id of > the original message, with a value that allows to distinguish between them. > > Regards, > Diego Gonzalez > Lagash Systems SA > > PS: Is this the correct mailing list to discuss about this submission?
Received on Wednesday, 9 November 2005 22:06:52 UTC