Re: [uddi-spec] Question regarding to tracking removed publisher assertionfor UDDI v3 subscription.

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

Alan,

Well, I think we definitely have something to fix here.   First off, the
"brief" attribute specifically states that the keyBag does not apply to
assertions.   Only problem with that is that there is no othre place to
specify something has been deleted.   I don't think we want to add an
asssertionStatusReport as a valid option to be in a keyBag - it is both
somewhat counter-intuitive and would affect uses of keyBag elsewhere in the
spec (e.g., in custody transfer).  The easiest thing may be to create an
entity specific to subscription which contains an assertionStatusReport and
an attribute indicating deletion.   We could alternatively add a deleted
attribute to the subscription structure (probably a better answer), but I
need to think through the ramificaitons of this a bit further I think.
let's discuss it at the FTF.

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


"Zhe (Alan) Wu" <Alan.Wu@oracle.com> on 10/21/2003 04:13:16 PM
STSM - Emerging Technologies
IBM Corporation

To:    "Rogers, Tony" <Tony.Rogers@ca.com>
cc:    Luc Clement <lclement@windows.microsoft.com>, UDDI Spec TC
       <uddi-spec@lists.oasis-open.org>
Subject:    Re: [uddi-spec] Question regarding to tracking removed
       publisher assertion for UDDI v3 subscription.



Hi Tony,

I guess we're talking about different things. I've no concern about
deletion
of publisher assertions. And I understand the part that you can view the
whole assertion as a unique key (a bit odd/inconsistent but acceptable).

Rather, after a publisher assertion is been deleted,
how is this 'change' been reflected in UDDI subscription result.

To make it a bit more concrete, consider a UDDI subscription established
using get_assertionStatusReport inquiry. If one assertionStatusItem is gone
due
to a removal of publisher assertion, how shoud we construct
subscriptionResultsList? Where do we show the fact that the
assertionStatusItem's
been removed? Do we put the whole assertionStatusItem in the keyBag, or
just
the assertion, ... etc.

Thanks,

Zhe (Alan) Wu
Oracle UDDI Server Team


Rogers, Tony wrote:
The bag inside a delete_publisherAssertions request isn't a keyBag, it's a
bag of publisher assertions.

When you delete a publisher assertion you don't do so by specifying its
key. You do so by specifying all the parts that make that publisher
assertion  unique - that's the two business keys and the relationship keyed
reference. I  suppose you could call the composite of those three things a
synthetic key, if  you wanted to, but that's stretching things a bit

Tony Rogers
tony.rogers@ca.com
-----Original Message-----
From: Zhe (Alan) Wu    [mailto:Alan.Wu@oracle.com]
Sent: Wed 22/10/2003 5:22
To: Luc Clement
Cc: UDDI Spec TC
Subject:    [uddi-spec] Question regarding to tracking removed publisher
assertion for    UDDI v3 subscription.


Dear Spec TC,

It seems that    UDDI v3 subscription API does address the changes
relevant to    publisherAssertion. My question is that if a publisher
assertion is    deleted, how can we put it inside the keyBag?
The xml schema does not    support that for the time being. However,
allowing a publisher assertion    reside in keyBag sounds weird too because
'keyBag' is supposed to be a bag    of keys and publisher assertion does
not
have a key.

Is it    reasonable to assign a key to publisher assertion, just to make it
consistent with other entities? That way, tracking deleted publisher
assertions
can also be done consistently.

Another way is to extend    the xml schema to allow keyBag include
publisher assertions. But then it    is better off to rename keyBag to
something else.

Thanks,

Zhe    (Alan) Wu
Oracle UDDI Server Team

The PDF and      HTML versions of the spec were updated and available
at:




HTML versions at:

   http://uddi.org/drafts/pubs/uddi_v3.htm
   http://uddi.org/drafts/pubs/uddi-v3.0.1-20031014.htm

PDF version at http://uddi.org/drafts/pubs/uddi-v3.0.1-20031014.pdf
Note      that: