Issue 2541: Datatypes: Expression (uml2-superstructure-ftf) Source: (, ) Nature: Uncategorized Issue Severity: Summary: Summary: the metaclass Expression includes an attribute called "language" of type Name. To enable tools to check OCL expressions, it is neccesary to define a standard value for this attribute, which denotes the fact that the expressions is an OCL expression. Without such a standard defined value tools cannot distinguish OCL expresions and cannot interpret them (for purposes of typechecking, code generation, etc....) I propose to add the value "OCL" as a standard value for the attribute "language" of metaclass "Expression" to the chapter on datatypes. Resolution: Revised Text: Actions taken: March 15, 1999: received issue March 9, 2005: closed issue Discussion: In UML 2.0 Infrastructure section 9.7.2, OpaqueExpresion, does state that the form ‘OCL’ should be used, though no specific values are defined for other languages: it is deferred to the source specification for that language. End of Annotations:===== From: jwarmer@nl.ibm.com X-Lotus-FromDomain: IBMNL@IBMGB To: uml-rtf@omg.org cc: sbrodsky@us.ibm.com, alan@trireme.com, Richard_Mitchell@nl.ibm.com, Bernard_Rumpe@nl.ibm.com, A.Kleppe@klasse.nl, sj_cook@uk.ibm.com Date: Mon, 15 Mar 1999 22:13:30 +0100 Subject: Datatypes: Expression Content-Disposition: inline Dear RTF, the metaclass Expression includes an attribute called "language" of type Name. To enable tools to check OCL expressions, it is neccesary to define a standard value for this attribute, which denotes the fact that the expressions is an OCL expression. Without such a standard defined value tools cannot distinguish OCL expresions and cannot interpret them (for purposes of typechecking, code generation, etc....) I propose to add the value "OCL" as a standard value for the attribute "language" of metaclass "Expression" to the chapter on datatypes. Regards, Jos ________________________________________________________________________ Jos Warmer IBM, Object Technology Practice OTP Leader Netherlands OCL home: http://www.software.ibm.com/ad/ocl private tel : +31 (0)35-6037656 work mobile: +31 (0)6 53428821 private fax : +31 (0)35-6037647 work email: jwarmer@nl.ibm.com http://www.klasse.nl/Engels/jos.htm From: "Griss, Martin" To: "'jwarmer@nl.ibm.com'" , uml-rtf@omg.org Cc: sbrodsky@us.ibm.com, alan@trireme.com, Richard_Mitchell@nl.ibm.com, Bernard_Rumpe@nl.ibm.com, A.Kleppe@klasse.nl, sj_cook@uk.ibm.com Subject: RE: Datatypes: Expression Date: Mon, 15 Mar 1999 14:32:00 -0800 Yes, good idea. Do we want version number too? OCL1.3 Martin L. Griss, Laboratory Scientist PS. Excuse terseness, errors - I continue with my RSI/tendonitis, and have started using a natural langauge system - less mistypes, more mispokes. HP Laboratories,1U-16 ofc=650-857-8715 1501 Page Mill Road, fax=650-813-3668 Palo Alto, CA 94301-1126 admin=650-857-7744 (Jennifer Jansen) http://www.hpl.hp.com/reuse > -----Original Message----- > From: jwarmer@nl.ibm.com [SMTP:jwarmer@nl.ibm.com] > Sent: Monday, March 15, 1999 1:14 PM > To: uml-rtf@omg.org > Cc: sbrodsky@us.ibm.com; alan@trireme.com; > Richard_Mitchell@nl.ibm.com; > Bernard_Rumpe@nl.ibm.com; A.Kleppe@klasse.nl; sj_cook@uk.ibm.com > Subject: Datatypes: Expression > > > > Dear RTF, > > the metaclass Expression includes an attribute called "language" of > type > Name. To enable tools to check OCL expressions, it is neccesary to > define a standard value for this attribute, which denotes the fact > that > the > expressions is an OCL expression. > Without such a standard defined value tools cannot distinguish OCL > expresions and cannot interpret them (for purposes of typechecking, > code generation, etc....) > > I propose to add the value "OCL" as a standard value for the > attribute > "language" of metaclass "Expression" to the chapter on datatypes. > > Regards, Jos > > > ________________________________________________________________________ > Jos Warmer IBM, Object Technology Practice > OTP Leader Netherlands OCL home: > http://www.software.ibm.com/ad/ocl > > private tel : +31 (0)35-6037656 work mobile: +31 (0)6 > 53428821 > private fax : +31 (0)35-6037647 work email: > jwarmer@nl.ibm.com > > http://www.klasse.nl/Engels/jos.htm > Subject: [issues 3391 and 2541] Classes Chapter Date: Fri, 12 Dec 2003 17:06:38 -0500 X-MS-Has-Attach: yes X-MS-TNEF-Correlator: Thread-Topic: [issues 3391 and 2541] Classes Chapter Thread-Index: AcPA/BabAACPi5DcQWWko7bCcHJvPg== From: "Karl Frank" To: X-OriginalArrivalTime: 12 Dec 2003 22:06:40.0068 (UTC) FILETIME=[38057840:01C3C0FC] As agreed in the FTF call this week, I am sending around some issues and proposed resolutions from the Classes chapter. Both issue 3391 and 2541 are to be proposed to be resolved by adding the quoted sentence below to the Semantics section for Opaque Expression. It should be noted that these issues were raised against the Expressions section, but that the late insertion of the OpaqueExpression appears to make OpaqueExpression the correct place for the correction. 'If the language attribute has the standard value "OCL", the body of the expression must be a syntactically correct and type correct OCL expression in the context of the Expression. The syntax and semantics of the expression is given by the OCL specification, currently OMG document number ptc/p3-10-14' Thanks to Anders Ivner for his help on this, and to Conrad Bock and an anonymous contributor for raising the issues. Feedback welcome. ------------- Karl Frank Borland Software Corporation cell: 978 853 3592 landline: 978 283 4656 15 Haskell Street Gloucester MA 01930 -------------------- The relevant parts of the spec for OpaqueExpression are: ================ OMG Issue No: 3391 Title: UML 1.4 RTF Issue: Multiple languages for uninterpreted strings Source: ObjectSwitch (Mr. Conrad Bock, conrad.bock@objectswitch.com) Summary: Multiple languages for uninterpreted strings The various places that uninterpreted strings are used in UML should support multiple languages. For example, the Expression metaclass has an metaattribute for language and another for the uninterpreted string. This should be a set of such pairs. Then code generators can target multiple languages from the same model. Discussion: {IF APPLICABLE - Summary of how the issue was proposed to be resolved and/or why it wasn't} Disposition: Unresolved ----------------------------------- OMG Issue No: 2541 Title: Datatypes: Expression Source: Summary: the metaclass Expression includes an attribute called "language" of type Name. To enable tools to check OCL expressions, it is neccesary to define a standard value for this attribute, which denotes the fact that the expressions is an OCL expression. Without such a standard defined value tools cannot distinguish OCL expresions and cannot interpret them (for purposes of typechecking, code generation, etc....) I propose to add the value "OCL" as a standard value for the attribute "language" of metaclass "Expression" to the chapter on datatypes. Discussion: {IF APPLICABLE - Summary of how the issue was proposed to be resolved and/or why it wasn't} Disposition: Unresolved Subject: InfraStructure Ballot 2 Issue 2541 Date: Mon, 26 Jan 2004 11:03:13 -0500 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Revised Framemaker files Thread-Index: AcPiD7i5Dv0LfiycSoy9szSdU4sA/wCDudAq From: "Karl Frank" To: Cc: , X-OriginalArrivalTime: 26 Jan 2004 16:03:30.0817 (UTC) FILETIME=[F1345F10:01C3E425] X-MIME-Autoconverted: from base64 to 8bit by amethyst.omg.org id i0QFvmRC019154 In my view, the "no change" resolution proposed for #2541 should be improved. For reference, the issue text is inserted below with the proposed no change disposition and supporting discussion. This is a shared issue of interest to the OCL and Super teams. OCL exists in at least two version now, the string "OCL" does not work to identify which parsing rules to use. - Karl Frank Borland Software Issue 2541: Datatypes: Expression (uml-rtf) Click here for this issue's archive. Nature: Uncategorized Issue Severity: Summary: Summary: the metaclass Expression includes an attribute called "language" of type Name. To enable tools to check OCL expressions, it is neccesary to define a standard value for this attribute, which denotes the fact that the expressions is an OCL expression. Without such a standard defined value tools cannot distinguish OCL expresions and cannot interpret them (for purposes of typechecking, code generation, etc....) I propose to add the value "OCL" as a standard value for the attribute "language" of metaclass "Expression" to the chapter on datatypes. Resolution: Revised Text: Actions taken: March 15, 1999: received issue Discussion: In UML 2.0 Infrastructure section 9.7.2, OpaqueExpresion, does state that the form â..OCLâ.. should be used, though no specific values are defined for other languages: it is deferred to the source specification for that language. Disposition: Closed, no change Subject: RE: Ballot 6 -- starts today at noon EST Date: Wed, 28 Jan 2004 09:40:13 -0500 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Ballot 6 -- starts today at noon EST Thread-Index: AcPlp9E5AkEQzswNTjSgorWW+Xl/2gAA919w From: "Karl Frank" To: "Branislav Selic" , Cc: X-OriginalArrivalTime: 28 Jan 2004 14:40:15.0327 (UTC) FILETIME=[A47C96F0:01C3E5AC] X-MIME-Autoconverted: from base64 to 8bit by amethyst.omg.org id i0SEYMJN018410 As Bran states below, Borland was unhappy with the resolution proposed for #2541. Although not critical, Borland will nevertheless vote No on that issue. The statement made in the resolution discussion is False. It is not the case that section 9.7.2 "does state that the form â..OCLâ.. should be used' The section instead notes that "One predefined language for specifying expressions is OCL". For those who understand the difference between use and mention of a word, it is clear that this statement does not say anything about what form to use in referencing this language, and that this is a parenthetical remark about what languages exist, not a normative statement about how conformant models shall designate that language. -Karl Frank | -----Original Message----- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Wed 1/28/2004 9:02 AM To: uml2-superstructure-ftf@omg.org Cc: mu2i-ftf@omg.org Subject: Ballot 6 -- starts today at noon EST Super FTF-ers, Enclosed is the official version of ballot 6. This vote is intended to endorse the resolutions already adopted by the MOF 2/UML Infrastructure FTF. The issues are shared infrastructure issues with relatively trivial solutions. (Note, however, that Karl was unhappy with one of the resolutions -- although I believe that he indicated that he did not feel that it was a critical issue for him.) If we do not endorse this ballot, the two FTFs will be out of sync and we will have a bit of a mess on our hands. However, notwithstanding this, you should check out the issues from a technical perspective and vote according to what you think is appropriate. The formal poll starts today and extends until Feb. 11. Because it is less than 2 weeks from the initiation of the previous vote (ballot 5), the rules for automatic expulsion from FTFs if you fail to vote get a bit complicated. To make my life easier, please cast your vote on ballot 6. And, of course, don't forget to cast your vote on ballot 5 as well, if you have not done so. There is one more week left for that ballot. Regards, Bran Selic Subject: RE: Ballot 6 -- starts today at noon EST Date: Wed, 28 Jan 2004 09:52:27 -0500 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Ballot 6 -- starts today at noon EST Thread-Index: AcPlp9E5AkEQzswNTjSgorWW+Xl/2gAA919wAACZ3xE= From: "Karl Frank" To: "Karl Frank" , "Branislav Selic" , Cc: X-OriginalArrivalTime: 28 Jan 2004 14:52:28.0421 (UTC) FILETIME=[5971DB50:01C3E5AE] X-MIME-Autoconverted: from base64 to 8bit by amethyst.omg.org id i0SEkTJN018619 Addendum: A 'Style Guideline' is non-normative. The mention of a string to use is properly shown by quotes or italics. When a correction is as simple as that required here, I cannot fathorm why it should not be made. - Karl -----Original Message----- From: Karl Frank Sent: Wed 1/28/2004 9:40 AM To: Branislav Selic; uml2-superstructure-ftf@omg.org Cc: mu2i-ftf@omg.org Subject: RE: Ballot 6 -- starts today at noon EST As Bran states below, Borland was unhappy with the resolution proposed for #2541. Although not critical, Borland will nevertheless vote No on that issue. The statement made in the resolution discussion is False. It is not the case that section 9.7.2 "does state that the form â..OCLâ.. should be used' The section instead notes that "One predefined language for specifying expressions is OCL". For those who understand the difference between use and mention of a word, it is clear that this statement does not say anything about what form to use in referencing this language, and that this is a parenthetical remark about what languages exist, not a normative statement about how conformant models shall designate that language. -Karl Frank | -----Original Message----- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Wed 1/28/2004 9:02 AM To: uml2-superstructure-ftf@omg.org Cc: mu2i-ftf@omg.org Subject: Ballot 6 -- starts today at noon EST Super FTF-ers, Enclosed is the official version of ballot 6. This vote is intended to endorse the resolutions already adopted by the MOF 2/UML Infrastructure FTF. The issues are shared infrastructure issues with relatively trivial solutions. (Note, however, that Karl was unhappy with one of the resolutions -- although I believe that he indicated that he did not feel that it was a critical issue for him.) If we do not endorse this ballot, the two FTFs will be out of sync and we will have a bit of a mess on our hands. However, notwithstanding this, you should check out the issues from a technical perspective and vote according to what you think is appropriate. The formal poll starts today and extends until Feb. 11. Because it is less than 2 weeks from the initiation of the previous vote (ballot 5), the rules for automatic expulsion from FTFs if you fail to vote get a bit complicated. To make my life easier, please cast your vote on ballot 6. And, of course, don't forget to cast your vote on ballot 5 as well, if you have not done so. There is one more week left for that ballot. Regards, Bran Selic Reply-To: Joaquin Miller X-Sender: jm-acm.no@sbcglobal.net@pop.sbcglobal.yahoo.com X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Tue, 10 Feb 2004 08:36:23 -0800 To: UML Superstructure FTF From: Joaquin Miller Subject: no votes X-Change Technologies votes no on these issues on Ballot 6: 2541: We will change our vote if the resolution is changed to specify that 'OCL 2.' is to be used to designate that OCL is the language used, where is replaced to complete the name of the OCL version used. 4446: We will change our vote if the resolution is changed to specify the effect of isStatic. X-Change Technologies votes yes on the other issues on Ballot 6. PGP Fingerprint: CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3 Cc: UML Superstructure FTF , cris.kobryn@telelogic.com From: Michael Latta Subject: More no votes on Ballot 6 Date: Tue, 10 Feb 2004 10:41:22 -0800 To: Joaquin Miller X-Mailer: Apple Mail (2.612) Ceira Technologies votes NO on the issues 2541 and 4446 for the reasons stated by X-Change Technologies and YES on all other issues in Ballot 6. We would like to see the resolution for 2541 include stated names for all OCL versions 1.x and 2.x. This is to allow legacy models to be imported into 2.x UML tools while retaining the accurate version of the OCL text being imported. It would be allowable to have tools that understand OCL auto-reformat the text, but tools that do not parse the OCL should leave the source version intact. The use of "OCL" without a version could be allowed as indicating a string asserted by the user to be OCL but not validated against one of the explicit versions. Michael Latta On Feb 10, 2004, at 8:36 AM, Joaquin Miller wrote: X-Change Technologies votes no on these issues on Ballot 6: 2541: We will change our vote if the resolution is changed to specify that 'OCL 2.' is to be used to designate that OCL is the language used, where is replaced to complete the name of the OCL version used. 4446: We will change our vote if the resolution is changed to specify the effect of isStatic. X-Change Technologies votes yes on the other issues on Ballot 6. PGP Fingerprint: CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3