Issue 604: Section 7.2: get_implementation function (orb_revision) Source: (, ) Nature: Uncategorized Severity: Summary: Summary: Why does Object have a get_implementation function instead of a readonly implementation attribute? (likewise for get_interface) Resolution: closed issue, no change Revised Text: Actions taken: July 9, 1997: received issue February 23, 1999: closed issue, no change Discussion: End of Annotations:===== Return-Path: To: orb_revision@omg.org Subject: testsuite issue Date: Wed, 09 Jul 1997 16:08:31 +0100 From: Stephen McNamara Section 7.2 Why does Object have a get_implementation function instead of a readonly implementation attribute? (likewise for get_interface) stephen Return-Path: X-Sender: vinoski@karloff.boston.iona.ie Date: Thu, 10 Jul 1997 10:35:14 -0400 To: Stephen McNamara From: Steve Vinoski Subject: Re: testsuite issue Cc: orb_revision@omg.org At 04:08 PM 7/9/97 +0100, Stephen McNamara wrote: >Section 7.2 > >Why does Object have a get_implementation function instead of a >readonly implementation attribute? (likewise for get_interface) Hi Stephen, 1) I believe we were supposed to have deprecated the Object::get_implementation() operation in the Portability Enhancement submission, but unfortunately the portability spec (orbos/97-05-1[56]) does not do so. I hope the Portability RTF remedies this oversight soon. I personally have considered get_implementation() to be a bug in the CORBA spec from day one, and have never implemented it in any ORB that I've ever helped develop -- no client should be able to get implementation details from an object! So all in all, I wouldn't worry too much about get_implementation(). 2) If get_interface() were instead defined as an attribute, it would look like this: interface Object { readonly attribute interface; ... }; This of course is illegal IDL, since the attribute name "interface" is already a keyword. --steve Return-Path: Sender: "Jeff Mischkinsky" Date: Thu, 10 Jul 1997 10:14:33 -0700 From: Jeff Mischkinsky Organization: Visigenic Software To: Steve Vinoski CC: orb_revision@omg.org Subject: Re: testsuite issue References: <3.0.32.19970709220538.006e0560@karloff.boston.iona.ie> Steve Vinoski wrote: > > At 04:08 PM 7/9/97 +0100, Stephen McNamara wrote: > >Section 7.2 > > > >Why does Object have a get_implementation function instead of a > >readonly implementation attribute? (likewise for get_interface) since attributes have no semantics except to define the "mapped operation name(s)" in the language it doesn't really matter. > > Hi Stephen, > > 1) I believe we were supposed to have deprecated the > Object::get_implementation() operation in the Portability > Enhancement > submission, but unfortunately the portability spec > (orbos/97-05-1[56]) does not do so. I hope the Portability RTF > remedies this oversight soon. I personally have considered > get_implementation() to be a bug in the CORBA spec from day one, and > have never implemented it in any ORB that I've ever helped develop > -- > no client should be able to get implementation details from an > object! So all in all, I wouldn't worry too much about > get_implementation(). have you filed this as an official issue for us to consider? > > 2) If get_interface() were instead defined as an attribute, it would > look like this: > > interface Object { > readonly attribute interface; > ... > }; > > This of course is illegal IDL, since the attribute name "interface" > is already a keyword. of course there is nothing that says you would have to name the attribute "interface". > > --steve -- Jeff Mischkinsky email: jeffm@visigenic.com Senior Software Architect voice: +1(415)312-5158 Visigenic Software, Inc. fax: +1(415)286-2475 951 Mariner's Island Blvd. Suite 460 web: http://www.visigenic.com San Mateo, CA 94404