[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [ws-caf-editors] Clarification on issue #136
That's not what was agreed to. There are protocols that have no concept
of nesting and should not have to have any reference to a parent, null
or not. Mark Little wrote: Yes, I understand. It's mandatory, but can have a null element ;-) Mark. ----- Original Message ----- From: "Simeon Greene" <simeon.m.greene@oracle.com> To: "Mark Little" <mark.little@arjuna.com> Cc: <ws-caf-editors@lists.oasis-open.org> Sent: Monday, August 09, 2004 3:46 PM Subject: Re: [ws-caf-editors] Clarification on issue #136This just came to mind. If it is mandatory, then the first Context in a chain will have to set it. And since that Context will not have a parent, what value would be used? Pointing to itself will be awkward (especially if the type is ContextType). Because of this I think we need to keep the element optional and change the text in the spec to reflect this. Regards, Simeon On Mon, 2004-08-09 at 06:08, Mark Little wrote:I thought the agreed motion was that the parent context was mandatory,butthat it could be a subset of the entire hierarchy. However, I waslisteningby phone at the time. Mark. ----- Original Message ----- From: "Simeon Greene" <simeon.m.greene@oracle.com> To: <ws-caf-editors@lists.oasis-open.org> Sent: Friday, August 06, 2004 8:45 PM Subject: [ws-caf-editors] Clarification on issue #136see http://services.arjuna.com/wscaf-issues/show_bug.cgi?id=136. The latest comments state: Change parent context from list to zero or one occurrence and further change parentsâ?T context to parentâ?Ts context. Alistair motioned Gregsecond.Noopposed. 8 in favor. It then goes on to say: begin with passed context must set the parent context. A strict interpretation of the text makes these two excerpts contradictory in my opinion. The former excerpt says the elementshouldbe optional "zero or one occurrence", yet the latter says that thebeginoperation "must" set it. This will cause a conflict between theschemaand the spec. I recommend we either change "must" to "may" or make parent context a mandatory element of the context. I prefer the first recommendation. |
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]