wsn message
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]
Subject: RE: [wsn] Topic/MessagePattern dialect
- From: Steve Graham <sggraham@us.ibm.com>
- To: "Patil, Sanjay" <sanjay.patil@sap.com>
- Date: Wed, 3 Nov 2004 08:29:50 -0500
Sanjay:
Yes, I was on a different point, but
I agree that Rick's original email was related to message pattern.
So, perhaps the best approach is to
amend the XSD definition of TopicExpression to include use="required"
on the @Dialect. That way, not specifying @Dialect yields a non-validating
TopicExpression.
sgg
++++++++
Steve Graham
(919)254-0615 (T/L 444)
STSM, On Demand Architecture
Member, IBM Academy of Technology
<Soli Deo Gloria/>
++++++++
"Patil, Sanjay"
<sanjay.patil@sap.com>
11/02/2004 02:52 PM
|
To
| Steve Graham/Raleigh/IBM@IBMUS
|
cc
| Rick Rineholt/Raleigh/IBM@IBMUS,
wsn@lists.oasis-open.org
|
Subject
| RE: [wsn] Topic/MessagePattern
dialect |
|
Hi Steve,
I thought the issue was about
the Dialect attribute on MessagePattern element that appears as part of
the Topic element in the TopicSpace. I am not sure if we have any message
exchanges associated with handling TopicSpace as such.
Consider the case where a NotificationProducer
obtains a TopicSpace from somewhere and wants to decide a particular topic
set to support. During this process, the NotificationProducer discovers
that the Dialect attribute on some of the MesssagePattern elements
in the TopicSpace are missing. Do we specify how to handle this situation?
If yes, how?
Thanks,
Sanjay
-----Original Message-----
From: Steve Graham [mailto:sggraham@us.ibm.com]
Sent: Tuesday, Nov 02, 2004 11:25 AM
To: Patil, Sanjay
Cc: Rick Rineholt; wsn@lists.oasis-open.org
Subject: RE: [wsn] Topic/MessagePattern dialect
One approach is to take this on a case by case basis. Currently TopicExpressions
are a component of the Subscribe Request, and we can specify in the description
of the Subscribe Request that a Dialect MUST appear, and the Subscribe
Request MUST fault if it is missing.
sgg
++++++++
Steve Graham
(919)254-0615 (T/L 444)
STSM, On Demand Architecture
Member, IBM Academy of Technology
<Soli Deo Gloria/>
++++++++
"Patil, Sanjay"
<sanjay.patil@sap.com>
11/02/2004 02:17 PM
|
To
| Rick Rineholt/Raleigh/IBM@IBMUS,
wsn@lists.oasis-open.org
|
cc
|
|
Subject
| RE: [wsn] Topic/MessagePattern
dialect |
|
Not sure what the specification could say here.
Ideally the Topics author should have specified a valid Dialect. If it's
not there, then perhaps there is a domain specific default in place, or
the author simply forgot to put the Dialect in there. I think the same
issue also applies when the Dialect value is garbage or it is simply not
recognized!
Providing an ability to specify default may offer some syntactic flexibility,
but it still can not guarantee full error handling.
Since there are no message exchanges associated in this case, the issue
can not be resolved by defining faults, etc.
Just some thoughts. Perhaps there are some good solutions that I am not
aware of.
Thanks,
Sanjay
>-----Original Message-----
>From: Rick Rineholt [mailto:rineholt@us.ibm.com]
>Sent: Tuesday, Nov 02, 2004 3:41 AM
>To: wsn@lists.oasis-open.org
>Subject: [wsn] Topic/MessagePattern dialect
>
>
> From my understanding of reading the wsdl/xsd, the MessagePattern
>element can contain any content and may have an optional Dialect
>attribute. The specification reads that the dialect(case seems
to be
>different) is to be used to essentially interpret the meaning of the
>content. From what I can find there is nothing that specifies when
the
>dialect attribute is not present how the content of
>MessagePattern is to
>be interpreted.
>
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]