Hi, sorry that I only just noticed this, but if we can have multiple
binding operation components with the same name, we need to update the
fragment identifier for these components to be able to address each
particular one of them. This is an issue important for the RDF mapping.

I cannot think at this moment of any distinguishing characteristic of
these same-named binding operations that can readily be used in the
component designators, though. And I thought that different bindings of
one operation can be done in different bindings, and these can then be
provided by the same endpoint, so the SPARQL WSDL can be refactored and
we can (again?) forbid multiple different binding operation components
with the same name.

I think we need email discussion before the f2f so that this can be
readily handled there.

Best regards,

Jacek

Thu, 2005-10-27 at 17:12 +0100, paul.downey (AT) bt (DOT) com wrote:
Hi Hugo,

I did mention this out in my review:

As discussed previously within this Working group, the service
is interesting in that the operation is bound to both HTTP and
SAP 1.2 over HTTP. In the case of HTTP, the operation is
bound twice using the whttp:method attribute to distinguish
between the GET and PST instances, both accepting input as
parameters.

I thought we had discussed binding the operation twice
to GET and PST during the Palo Alto (TIBC) F2F, and decided
it was K, but I'm now questioning my memory and the WSDL 2.0
model in my head. Sorry, I should have flagged this up more
strongly during my walk-through on last week's call.

Paul

Message
From: Hugo Haas [mailto:hugo (AT) w3 (DOT) org]
Sent: Thu 10/27/2005 12:08 PM
To: Downey,P,Paul,CXMA C
Cc: jmarsh (AT) microsoft (DOT) com; www-ws-desc (AT) w3 (DOT) org
Subject: Re: Attempted review of SPARQL Protocol LC Draft

Hi Paul.

Sorry to comment late on your review. However, I noticed something in
the draft:

* paul.downey (AT) bt (DOT) com <paul.downey (AT) bt (DOT) com[2005-10-19 16:25+0100]
The draft calls attention (in red text) to three WSDL 2.0
issues raised by this use-case:
- the requirement to have a single output media type,
- and a single fault media type
which we recorded as LC337 and LC338 respectively:
#LC337
#LC338
- the inability of having an inputSerialization of
"application/x-www-urlencoded" when the value a binding style
is "", which we
recorded as:
#LC345

I noticed the following in the draft:

<binding name="queryHttp" interface="tns:SparqlQuery"
type=""
whttp:version="1.1">

<fault name="MalformedQuery" whttp:code="400"/>
<fault name="QueryRequestRefused" whttp:code="500"/>

<!-- the GET binding for query operation
<operation ref="tns:query" whttp:method="GET"
whttp:inputSerialization="" />

<!-- the PST binding for query operation
<operation ref="tns:query" whttp:method="PST"
whttp:inputSerialization="" />

</binding>

Can one bind an operation twice? It was not clear from my reading of
the specification whether it was allowed.

If this isn't the case, do we have an issue about this from the DAWG?
If not, we should make this comment to the DAWG, and figure out what
to do here.

As Kendall raised a number of issues, I have lost track of where we
are with accomodating the SPARQL Protocol use of WSDL 2.0.

Regards,

Hugo

Fri, 2005-11-04 at 11:51 +0100, Hugo Haas wrote:
I cannot think at this moment of any distinguishing characteristic of
these same-named binding operations that can readily be used in the
component designators, though. And I thought that different bindings of
one operation can be done in different bindings, and these can then be
provided by the same endpoint, so the SPARQL WSDL can be refactored and
we can (again?) forbid multiple different binding operation components
with the same name.

I'm wondering if this means that we are going to require in this case
a name property for binding operations to distinguish them.

Wouldn't this be confusing?

interface/operation name="getStockQuote"
binding/operation ref="getStockQuote" name="somethingElse"?

Instead, I suggest again that we require up to one binding operation for
a single interface operation, as it seems that two different bindings
with the same endpoint address would do the job equally well for SPARQL.

In simple words - if you want to bind an operation twice and
differently, use two bindings.

Best regards,

Jacek

Fri, 2005-11-04 at 11:55 +0100, Jacek Kopecky wrote:

In simple words - if you want to bind an operation twice and
differently, use two bindings.

+1.

In fact, I don't see how our component model supports having two binding
operation components with the same name.

Sanjiva.