Issue 2495: MOF is using CORBA string for its string data types (mof-rtf) Source: (, ) Nature: Enhancement Severity: Minor Summary: Summary: MOF is using CORBA string for its string data types. It should use a wide string. UNICODE is needed to support widespread interoperability. Resolution: close with no further action Revised Text: Actions taken: March 1, 1999: received issue December 3, 2001: closed issue Discussion: This issue has been addressed by the resolutions to Issues 2198 and 4351. All string types in the MOF Model are now wide-strings End of Annotations:===== Date: Sat, 27 Feb 1999 09:50:21 +0100 From: Mario Zannone Organization: Genesis Development srl X-Accept-Language: en To: mof-rtf@omg.org CC: David Frankel , Werner Froidevaux Subject: MOF to IDL mapping for Attributes and References Hi, Why is the IDL template for References different from the one for Attributes? 1.--------- add_ operations raise Reflective::SemanticError, while add_ operations do NOT raise Reflective::SemanticError. 2.--------- {add|modify|remove}_at operations are generated for Attributes but not for References. Thanks, /Mario To: Mario Zannone cc: mof-rtf@omg.org, David Frankel , Werner Froidevaux , crawley@dstc.edu.au Subject: Re: MOF to IDL mapping for Attributes and References Date: Sun, 28 Feb 1999 23:10:23 +1000 From: Stephen Crawley Mario, You wrote: > Why is the IDL template for References different > from the one for Attributes? > > 1.--------- > add_ operations raise Reflective::SemanticError, > while > add_ operations do NOT raise > Reflective::SemanticError. That is a bug. The reference "add" operations should have raised SemanticError. However, that is largely academic now ... The RTF has recommended that exceptions raised by operations in the MOF specific and generic (Reflective) interfaces should be overhauled. In particular, we are rolling SemanticError, StructuralError and ConstraintError into a single error. [See issue MOF RTF 1085] > 2.--------- > {add|modify|remove}_at operations are generated for Attributes > but not for References. These operations are unnecessary for References. Unlike Attributes, the "collections" that the user sees via a Reference are guaranteed to have no duplicates ... because Associations don't allow duplicates. -- Steve To: Mario Zannone Cc: mof-rtf@omg.org, David Frankel , Werner Froidevaux Subject: Re: MOF to IDL mapping for Attributes and References Date: Mon, 01 Mar 1999 10:22:40 +1000 From: Kerry Raymond > add_ operations raise Reflective::SemanticError, while > add_ operations do NOT raise Reflective::SemanticError. That's a mistake and is already fixed in the RTF (from memory). > {add|modify|remove}_at operations are generated for Attributes > but not for References. This is a more interesting question as Attributes and References were supposed to be treated similarly (a design decision). However, as I recall, when the MOF submission was due, we were still having problems getting resolution on UML alignment issue of the semantics of ordering of associations. In the MOF, an ordered multi-valued attribute was well-understood semantically. However, references are just a view of the association, and there were concerns about the possible view-update implications on the associations where both association ends were ordered. So, the operations were not generated for references. (Admittedly, I think the same problem occurs with the set_ operation, which is generated, so maybe we only half-finessed the problem in the first place). Currently we still have a number of open issues in the MOF RTF which relate to the changeability, navigability, and (now) orderability of UML associations. Getting some precise semantics for these would be very useful. Kerry ============================================================================= Dr Kerry Raymond, Architecture Unit Leader E-mail: kerry@dstc.edu.au CRC for Distributed Systems Technology Phone: +61 7 3365 4310 University of Queensland 4072 Australia Fax: +61 7 3365 4311 =========================================== WWW: http://www.dstc.edu.au/kerry To: Mario Zannone , mof-rtf@omg.org, David Frankel , Werner Froidevaux Subject: Re: MOF to IDL mapping for Attributes and References Date: Mon, 01 Mar 1999 11:01:49 +1000 From: Kerry Raymond > (Admittedly, I think the same problem occurs with the set_ > operation, which is generated, so maybe we only half-finessed the problem > in the first place). I should have realised when I wrote this that my reasoning was wrong. Steve's explanation for the lack of add/mofify/remove_*_at is the correct one. Kerry ============================================================================= Dr Kerry Raymond, Architecture Unit Leader E-mail: kerry@dstc.edu.au CRC for Distributed Systems Technology Phone: +61 7 3365 4310 University of Queensland 4072 Australia Fax: +61 7 3365 4311 =========================================== WWW: http://www.dstc.edu.au/kerry X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4 To: mof-rtf@omg.org Subject: One more issue resolution for Block 3 Mime-Version: 1.0 Date: Fri, 06 Jul 2001 12:25:33 +1000 From: Stephen Crawley X-Scanned-By: MIMEDefang 1.0 (http://www.roaringpenguin.com/mimedefang/) Content-Type: text/plain; charset=us-ascii X-UIDL: jBW!!<^:!!?N/e9=2G!! I just realised that we can close Issue 2495 because we've addressed it in 2198. This issue can now be resolved with "close no action". > Issue 2495: MOF is using CORBA string for its string data types > > Nature: Enhancement > Severity: Minor > Summary: MOF is using CORBA string for its string data types. It > should use > a wide string. UNICODE is needed to support widespread > interoperability. Proposed Resolution: This issue has been addressed by the resolutions to Issues 2198 and 4351. All string types in the MOF Model are now wide-strings. Close with no further action. Proposed Revised Text: none