Re: [uddi-spec] Possible omission?

[Tom Bellwood <bellwood@us.ibm.com>]

Tony,

Sorry for not posting on this earlier.   These look like a few holes we can
close.  I'm adding your comments as an agenda item to the 7/8 telecon, to
be discussed for consideration as either an add to CR002, or as a stand
alone CR.

Thanks,
Tom Bellwood       Phone:  (512) 838-9957 (external);   TL:  678/9957
(internal)
Co-Chair, OASIS UDDI Specification TC


"Rogers, Tony" <Tony.Rogers@ca.com> on 06/24/2003 01:46:27 AM
STSM - Emerging Technologies
IBM Corporation

To:    <uddi-spec@lists.oasis-open.org>
cc:
Subject:    [uddi-spec] Possible omission?




I was just re-reading the UDDI V3 spec, as one does from time to time, and
I suspect I may have spotted an omission.

Under 5.2.17 save_service we have:




5.2.17.3 Behavior:


Each businessService passed MUST contain a businessKey value that
corresponds to a

registered businessEntity controlled by the same publisher who is  making
the save_service

request.



But there is no statement under any of the possible error codes  indicating
which error code should be thrown if the request does NOT contain a
businessKey value.

We know that an erroneous business key will result in  E_invalidKeyPassed,
but is that appropriate when the business key is missing  from the request?
Even so, it is probably the best choice of the ones  available.

I suggest we add the following sentence under  E_invalidKeyPassed: "This
also applies when a businessService element is passed  without the required
businessKey.".



We will need to make a similar change to save_binding.



While looking at this, I realised that the sentence I quoted above  (the
first sentence under 5.2.17.3) is slightly wrong, too. Shouldn't it be
qualified with: "unless the serviceKey is specified and denotes an existing
businessService entry in the repository."? In other words, we don't need
the  business key if we already know about the service?



Sorry to be a pain.

Tony Rogers

tony.rogers@ca.com