Re: Optional attributes



Date view Thread view Subject view Author view Attachment view

From: Jos Warmer (J.Warmer@klasse.nl)
Date: Tue 18 May 2004 - 21:32:09 BST


Shane, John,

Any issues with the OCL 2.0 specification should be sent to the official 
OMG address: issues@omg.org.
Last week I was told that the current planning for finalization of all of 
the UML 2.0 specifications is
targeted for september this year. Therefore there is enough time to make 
changes before that.
After finalization there is always a Revision TaskForce (RTF) that takes up 
these matters.

At this moment in time I am not involved in OMG work directly and therefore 
also not in the finalization
of the OCL 2.0.  I am sure that the original submission team (including 
myself) will be consulted when
neccesary, but please make sure you send all your remarks through the 
official OMG channels. I do
appreciate it, however, if you CC me.

Jos

At 09:56 PM 5/18/2004 +0200, Shane Sendall wrote:
>John,
>I don't really know. It is probably possible to send it directly to the 
>submission team or formally through the OMG list. I would certainly try to 
>send it to the submission team first. I have cc-ed Jos. He will at least 
>know the right person to talk to if that isn't him even. I think there 
>still should be time for changes before finalization, but I'm not sure of 
>the schedule. I guess they will just need to agree with our analysis of 
>the situation and eventually the proposal.
>I would nevertheless be still for an additional 'oclIsNull' operation 
>(especially as it is basically syntactic sugar and hence doesn't require 
>much effort to add). As you say 'isEmpty' is inelegant and even misleading IMO
>
>Jos, I have also some other suggestions. Can you keep me in the loop (wrt 
>your response)?
>
>Cheers,
>     Shane
>
>John Daniels wrote:
>>
>>Hi Shane,
>>
>>
>>>
>>>With this approach, "null" values can still be represented for entities
>>>with a [0..1] multiplicity constraint. In this case, we could have an
>>>operation "oclIsNull", which returns true if the result is empty, and
>>>false otherwise, making use of OCL's double interpretation
>>>(object/collection) of navigation to entities with [0..1].
>>>
>>
>>
>>We already have isEmpty() and notEmpty() which, while inelegant, do
>>the job. They are what I've always used.
>>
>>Do I need to submit something formally to someone to get this whole
>>issue looked at during the finalization?
>>
>>--John
>>
>>
>
>
>_____________________________________________________
>Klasse Objecten         tel     : +31 (0)35 6037646
>Chalonhof 153           fax     : +31 (0)35 6037647
>3762 CT Soest           email   : J.Warmer@klasse.nl
>The Netherlands         internet: http://www.klasse.nl

Date view Thread view Subject view Author view Attachment view