Issue 753: Inheriting exceptions in IDL (orb_revision) Source: (, ) Nature: Uncategorized Issue Severity: Significant Summary: Summary: When writing IDL, why is it not possible to have an exception declaration inherit from another exception? It"s possible to have an interface inherit another interface, so it seems logical to derive an exception class from an already declared exception area Resolution: close issue with no change with the above explanation. Revised Text: Actions taken: October 23, 1997: received issue February 23, 1999: closed issue Discussion: End of Annotations:===== Return-Path: X-Sender: andrew@emerald.omg.org Date: Thu, 23 Oct 1997 11:47:27 -0400 To: Juergen Boldt From: Andrew Watson Subject: IDL Exceptions Juergen, Please could you file this as an issue for the Core RTF, under the title "Inheriting exceptions in IDL"? Thanks, Andrew From: jmmelton@bkb.com To: " - (052)andrew(a)omg.org" Cc: " - (052)mark.brubaker(a)raleigh.hcl.com" Subject: IDL Exceptions Date: Mon, 20 Oct 1997 15:27:43 -0400 Andrew, I noted the discussion below in the OMG archive for "Ask the Experts" on the Web. I would like to add my voice to those who feel inheriting from a base exception class makes managing exceptions much easier for development. Although the difficulties in implementing an exception inheritance heirarchy are noted for non-object oriented languages, I assume similar difficulties were overcome for interface inheritance in those languages. BTW: In answer to your question to Mark, I am developing in C++ and Java. Thanks, James Melton Global Capital Markets Systems, BankBoston > Re: IDL Exceptions > > ---------------------------------------------------------------------------- > > * To: mark.brubaker@raleigh.hcl.com > * Subject: Re: IDL Exceptions > * From: Andrew Watson > * Date: Tue, 1 Jul 1997 19:35:05 -0400 > * Cc: experts@omg.org > * In-Reply-To: <199706251821.OAA26514@emerald> > > ---------------------------------------------------------------------------- > > Mark, > > You wrote: > > > Key: Wed Jun 25 14:21:32 EDT 1997: 541.534423828125 > > > Questions: When writing IDL, why is it not possible to > > have an exception declaration inherit from > > another exception? It is possible to have an > > interface inherit another interface, so it > > seems logical to me (and I have the need) to > > derive one exception class from an already > > declared exception class. > > It's an interesting suggestion. Since exceptions are usually so simple and > one so rarely creates one that's an extension of another, the CORBA > designers probably didn't think it would save you much time to be able > inherit from one. > > What programming language are you using? Although languages like C++ > support hierarchies of exception handlers, other languages with IDL > mappings do not (C & COBOL come to mind), so this might actually hinder > creating IDL mappings for these languages. > > Suggestions and comments on OMG specs can be sent issues@omg.org - we file > them and pass them on to the relevant revision task force for consideration > when making the next revision of a specification (but note that this can > take up to a year). > > Regards, > > Andrew > Andrew Watson Tel: +1 508 820 4300 > Object Management Group, Fax: +1 508 820 4303 > 492 Old Connecticut Path Email: andrew@omg.org > Framingham, MA 01701, USA http://www.omg.org/~andrew Return-Path: Sender: Bill Janssen From: Bill Janssen To: orb_revision@omg.org Subject: orb_revision issue 753: exception inheritance in IDL Date: Thu, 9 Jul 1998 13:20:05 PDT I don't believe the justification given for the resolution is accurate. First, no consideration of single inheritance is given, which would be substantially simpler to support (and still quite useful) than would be multiple inheritance. Secondly, it's not all that tough to imagine good mappings of a multiple inheritance exception system to languages like C which don't support inheritance. Since in C we already manipulate the values of the exception through formalized accessors, any "messy" details of the mapping could be satisfactorily hidden from application code. I believe that the real resolution for this issue should be that this change is too big for an RTF to make, and therefore will not be considered further in this venue. Bill Return-Path: Sender: jis@fpk.hp.com Date: Thu, 09 Jul 1998 17:18:28 -0400 From: Jishnu Mukerji Organization: Hewlett-Packard New Jersey Laboratories To: Bill Janssen Cc: orb_revision@omg.org Subject: Re: orb_revision issue 753: exception inheritance in IDL References: <98Jul9.132005pdt."404627"@watson.parc.xerox.com> Bill Janssen wrote: > I don't believe the justification given for the resolution is accurate. > > First, no consideration of single inheritance is given, which would > be substantially simpler to support (and still quite useful) than > would be multiple inheritance. > > Secondly, it's not all that tough to imagine good mappings of a > multiple inheritance exception system to languages like C which don't > support inheritance. Since in C we already manipulate the values of > the exception through formalized accessors, any "messy" details of > the mapping could be satisfactorily hidden from application code. > > I believe that the real resolution for this issue should be that > this change is too big for an RTF to make, and therefore will not > be considered further in this venue. > > Bill OK, if there are no objections I will add that as a further annotation. The bottom line still remains that the issue is close no change. Thanks, Jishnu.