Re: [wsrf] [WSRF 4] continued discussion

[Tim Banks <tim_banks@uk.ibm.com>]

This issue is unresolved; let's see if we have a conclusion.

Steve Graham <sggraham@us.ibm.com> wrote on 20/09/2004 13:44:37:

>
> I need to propose some text to resolve Issue wsrf 4.
>
> Here is the text of the issue:
> Issue WSRF4: GetResourceProperty on property document with xsd:any
> When a property document schema contains an xsd:any element and a
> GetResourceProperty message is received for a property QName for
> which there is currently no instance, what should be the response?

At the F2F in July, three cases were identified:

a) Where the property is defined, but minoccurs=0.  We agreed a 'nil'
response is right.
b) For an instance document with a dynamic ‘stacktrace’ property,  a
request for ‘stooktrace’ should return an error.
c) What if there is a request for a property that sometimes is present?

We observed that: the server may behave as if it’s there, or if it's not.
It may even depend on the timing of the request vs the creation of the
property by the service, so  we can only describe what the service could
do.

Steve Proposed (on 20th Sept) a modification of 'section 5.1, defining
GetResourceProperty message exchange'  of the RP spec to allow for dynamic
properties:
>
> /wsrp:GetResourceProperty/QName
> This MUST correspond to the QName of a resource property element
> child of the root of the WS-Resource’s resource properties document.
>  Removing the word "defined" eliminates the possible interpretation
> that the QName must be defined in the schema of the RP document.

Which got +1 from Tom Maguire, Brian Murray and (now) myself.


So, I think the current proposal is that the server may respond with a
fault OR a 'nil' for dynamic properties, but should not respond with 'nil'
to requests which are misspelled. The client may use
GetAllResourceProperties request (See issue 27) to discover, at any
particular instant, the full range of property names and has a very good
chance of detecting misbehaviour.  This advice should go in the AppNotes
document.

Issue 4 can be moved to 'resolved' on this basis, I think.

This solution could be further refined following the outline agreed at the
July f2f around the issue now known as WSRF 66:

Proposed a) A new resource property – allowedRPQNames, ie those which will
not throw ‘unknown resource’ fault.
This properties will return the relevant RP QNames and the schemalocation.
(No objections)
Proposed b) A new resource property – presentRPQNnames, which indicates a
property of the relevant name in the instance document.
This property will return the relevant RP QNames and the schemalocation.
(No Objections)
Proposed c) A capability - GetRPDecl (QName) -  to return the current RP
document declaration ( minoccurs/maxoccurs/nillable etc) for each QName
requested.
(No Objections)