Issue 8154: Section: 11.3.10 (uml2-rtf) Source: (, ) Nature: Clarification Severity: Minor Summary: Add OCL notation to Constraints. Typo - place a period at the end of the definition of the operation:Operation[1] association Add a subsets or specialization statement to the association target:InputPin[1] as is shown in fig. 144 Resolution: OCL is duplicate with 6346 Revised Text: In Actions, CallOperationAction, Associations operation entry, add period at end of description. Editor’s note: fixed in formal copy edit target entry, at end of description, add “(subsets Action::input)”. Actions taken: January 27, 2005: received issue August 23, 2006: closed issue Discussion: End of Annotations:===== m: webmaster@omg.org Date: 27 Jan 2005 15:06:36 -0500 To: Subject: Issue/Bug Report -------------------------------------------------------------------------------- Name: Jane Messenger Company: U. S. Geological Survey mailFrom: jmessenger@usgs.gov Notification: Yes Specification: Superstructure Section: 11.3.10 FormalNumber: ptc/04-10-02 Version: 2.0 Draft Adopted RevisionDate: 10/08/2004 Page: 262 Nature: Clarification Severity: Minor HTTP User Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0; Q312461) Description Add OCL notation to Constraints. Typo - place a period at the end of the definition of the operation:Operation[1] association Add a subsets or specialization statement to the association target:InputPin[1] as is shown in fig. 144 Subject: RE: ,ac,,av, AA WG discussion for ballot 9, take 2 Date: Mon, 22 Aug 2005 10:31:34 +0200 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: ,ac,,av, AA WG discussion for ballot 9, take 2 thread-index: AcWmeu2rrpR3yss5R/eaAeKHiFv/pAAd7jHw From: "Tim Weilkiens" To: , "uml2rtf" X-MIME-Autoconverted: from quoted-printable to 8bit by amethyst.omg.org id j7M8mfhh013245 Conrad, I've did some OCLs at the beginning of the RTF, but didn't finish due some open issues. I have the OCLs for 8154: Add OCL expressions for constraints 1,2, and 4: [1] The number of argument pins and the number of owned parameters of the operation of type in and in-out must be equal. ((self.operation.ownedParameter->select(par | par.direction=#in or par.direction=#inout))->size()) = self.argument->size() [2] The number of result pins and the number of owned parameters of the operation of type return, out, and in-out must be equal. self.operation.returnResult()->size() = self.result->size() [4] The type of the target pin must be the same as the type that owns the operation. (self.operation.class->notEmpty() implies self.operation.class = self.target.type) and (self.operation.datatype->notEmpty() implies self.operation.datatype = self.target.type) and (self.operation.artifact->notEmpty() implies self.operation.artifact = self.target.type) Note that OCL constraint [4] depends on open issues 8692 und 8693. Here are OCLs for 8192: [1] self.object.type = self.unmarshallType [2] (self.object.lower = 1) and (self.object.upper = 1) [3] self.result->size() = self.unmarshallClassifier.feature->select(f | f.isTypeOf(StructuralFeature))->size() [4] structural features have no order Tim > -----Original Message----- > From: Conrad Bock [mailto:conrad.bock@nist.gov] > Sent: Sunday, August 21, 2005 8:01 PM > To: uml2rtf > Subject: ,ac,,av, AA WG discussion for ballot 9, take 2 > > > Internal WG discussion only. > Bran: These are not proposals for ballot (yet). > > AA WG, > > Some of of the resolutions I proposed for ballot 9 discussion were > already done, thanks to Bran for pointing that out. Here is > an update. > If you have another others, send them to me by Wednesday COB. > > Conrad Subject: RE: Draft ballot 9 Date: Sun, 28 Aug 2005 21:14:28 -0700 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Draft ballot 9 Thread-Index: AcWqh4MNCYn98KMqRxSBzpBmZwi15QBw8hIQ From: "Pete Rivett" To: "Branislav Selic" , 8154: very minor typo on last line of resolution: 'ad end of description' 8154: the resolution inserts text .(subsets Action:input). - should that not be .(subsets Action::input). (double colon)? If so the same change is needed in several other of the resolutions. 8192 - missing close parenthesis in: 'Generalization, replace .AcceptEventAction (from CompleteActions). with .Action (from BasicActions..' 8208 - the addition to diagram would benefit from multiplicities. Since it has no arrow on the line is it the case that both ends will be owned by the respective classes? If so there should be a name at both ends. 8746: In the case of a 'multi party protocol' I would exepct a class representign the protocol itself whoch would have the associatiosn to the interfaces for the roles. 8939 The issue refers to LoopNode as well as ConditionalNode needing this constraint. Should really have OCL for the constraint - as worded it seems somewhat open to interpretation. PS while looking at this I noticed that in section 12.3.40 there is no Generalizations section for OutputPin. Has this been picked up alreayd? Pete Pete Rivett (mailto:pete.rivett@adaptive.com) CTO, Adaptive Inc. Dean Park House, 8-10 Dean Park Crescent, Bournemouth, BH1 1HL, UK Tel: +44 (0)1202 449419 Fax: +44 (0)1202 449448 http://www.adaptive.com -------------------------------------------------------------------------------- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Friday, August 26, 2005 10:45 PM To: uml2-rtf@omg.org Subject: Draft ballot 9 Here is the first draft of ballot 9 for soaking over the next week. It is the skinniest one we've ever put out (just 22 proposals), which is a bit embarassing for a team that nominally consists of 26 members. In my panic about this, I threw in 3 additional resolutions that I had not published before, so please review them carefully: 8696 8728 8746 Here are the changes from previously published versions of these same resolutions: 8066 -- accounted for Pete's and Conrad's comments and simplified the proposed text additions. 8084 -- Kenn pointed out that this problem was already resolved by the FTF. However, it turns out that the FTF resolution was not applied to the Infrastructure, even though it was supposed to be. So, I added the change to the infrastructure (perhaps this fix could be added to the "formal" version of the Infrastructure spec that Linda is working on). 8097 -- added a comma as per Conrad's recommendation (I actually disagree that this comma is necessary but decided that I cannot win any arguments with Conrad. I do not believe that every "which" should be preceded by a comma -- no matter what Microsoft Word says) 8100 -- Changed to reflect Pete's suggestions. (I too agree that the different language strings are meant to be alternatives, but I guess something in the way I wrote the "clarifications" text made Pete think otherwise, so I modified the text) 8226 -- Changed the BNF along the lines that Pete suggested with one modification (to prevent empty "{ }" pairs). However, even so, the BNF is still not tight enough and needs a constraint since it still allows duplicates such as {ordered, ordered}. But, this type of constraint cannot be expressed in BNF. I don't think it's an improvement, since the original problem could also have been solved with a constraint, but since I cannot seem to win any arguments with Pete, I gave in. 8719 -- removed the confusing example as per Pete's sugesstion and added a piece of clarifying text as per Conrad's suggestion. Please provide your feedback on these resolutions before the ballot is formally issued next Friday. Objections after that point are problematic and cause confusion. Cheers, Bran To: "Pete Rivett" Cc: uml2-rtf@omg.org Subject: RE: Draft ballot 9 X-Mailer: Lotus Notes Release 6.0.1CF1 March 04, 2003 From: Branislav Selic Date: Mon, 29 Aug 2005 08:23:21 -0400 X-MIMETrack: Serialize by Router on D25ML01/25/M/IBM(Release 6.5.4|March 27, 2005) at 08/29/2005 08:23:23, Serialize complete at 08/29/2005 08:23:23 Pete, Thanks a lot for the detailed review of the draft resolutions, I will make the editorial fixes you've identified. A few comments: > 8154: the resolution inserts text .(subsets Action:input). - should > that not be .(subsets Action::input). (double colon)? If so the same > change is needed in several other of the resolutions. It should be double colon -- thanks for catching this. Actually, we have a number of different ways of specifying this information, and the change should really be made consistently throughout the spec. This inconsistency should probably be raised as a separate issue. Coincidentally, the resolution to issue 8696 actually addresses the issue of the preferred format. > 8746: In the case of a 'multi party protocol' I would exepct a class > representign the protocol itself whoch would have the associatiosn > to the interfaces for the roles. That is one possibility, but this enters into the realm of design choices. Another is to define the protocol using a collaboration. I don't know how far we should go into such things in the spec. (In fact, I actually don't like having the example in the spec in the first place, but that is just my opinion.)