Return-Path: Delivered-To: apmail-xml-axis-dev-archive@xml.apache.org Received: (qmail 24823 invoked by uid 500); 16 Nov 2001 14:11:09 -0000 Mailing-List: contact axis-dev-help@xml.apache.org; run by ezmlm Precedence: bulk Reply-To: axis-dev@xml.apache.org list-help: list-unsubscribe: list-post: Delivered-To: mailing list axis-dev@xml.apache.org Received: (qmail 24775 invoked from network); 16 Nov 2001 14:11:08 -0000 Message-ID: <003901c16ea8$700ebe20$9fa17cca@watson.ibm.com> From: "Sanjiva Weerawarana" To: References: <47FD4196C273D411950300508BCF197801656D06@S0001EXC0006> <032501c16e47$328cb4a0$48a17cca@watson.ibm.com> <001b01c16e4a$a9ff9860$b3185480@genesis.adtech.internet.ibm.com> Subject: Re: What do we do with operation types: notification, solicit-response? Date: Fri, 16 Nov 2001 20:10:03 +0600 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N What are the semantics of solicit-response and notification operations? If I'm a service S with solicit-response operation o1 of port type pt1, from whom do I solicit a response? Using instant messaging as a transport is an orthogonal issue I believe. It should be possible to implement SOAP RPC using an IM transport such as Jabber (and I believe such code exists). Sanjiva. ----- Original Message ----- From: "Benjamin Szekely" To: Sent: Friday, November 16, 2001 8:59 AM Subject: Re: What do we do with operation types: notification, solicit-response? > I've been working on how to implement solicit-response and notification > services using instant messaging as the transport. The way I do it is I > create a custom Java Provider which passes IM identities and pressence > (awareness) information to the service object. I have some demos of this > that I did for soap 2.2. I have two different IM transports for Axis which > I could easily write providers for. > > Ben > > ----- Original Message ----- > From: "Sanjiva Weerawarana" > To: > Sent: Thursday, November 15, 2001 9:33 PM > Subject: Re: What do we do with operation types: notification, > solicit-response? > > > > +1 > > > > The semantics of solicit-response and output-only are not quite > > fully defined. I'd rather have the semantics fixed properly > > rather than do any implementation of it. > > > > Sanjiva. > > > > ----- Original Message ----- > > From: "Tom Jordahl" > > To: > > Sent: Friday, November 16, 2001 12:03 AM > > Subject: RE: What do we do with operation types: notification, > > solicit-response? > > > > > > > > > > I would say generating a warning (and adding an option to suppress > > warnings) > > > and ignoring the operation would be a reasonable thing to do. > > > > > > This would allow someone to use a WSDL file which had one-way operations > > in > > > it in addition to the request-response operations that they wanted to > use. > > > > > > -- > > > Tom Jordahl > > > > > > > > > -----Original Message----- > > > From: Russell Butek [mailto:butek@us.ibm.com] > > > Sent: Thursday, November 15, 2001 12:59 PM > > > To: axis-dev@xml.apache.org > > > Subject: What do we do with operation types: notification, > > > solicit-response? > > > > > > > > > WSDL defines 4 operation types: > > > request-response: in message, out message, fault message optional > > > one-way: in message > > > solicit-response: out message, in message, fault message optional > > > notification: out message > > > > > > JAX-RPC says "JAX-RPC 1.0 specification supports the mapping of > operations > > > with request-response and one-way transmission primitives. The standard > > > Java mapping of operations defined with other transmission primitives > > > (notification, solicit-response) is considered out of scope in the > JAX-RPC > > > 1.0 specification." > > > > > > Earlier we had decided to throw an exception if Wsdl2java encounters a > > > notification or solicit-response operation. But now that I'm thinking > > > about implementing this check, I have a question. Do we complain and > > > ignore the operation, generating everything else? Or do we fail > outright > > > and not generate anything? > > > > > > Russell Butek > > > butek@us.ibm.com > >