Issue 2937: CASE troubles (pids-rtf2) Source: Cognition Group, Inc. (Dr. David Forslund, nobody) Nature: Uncategorized Issue Severity: Summary: Since with CORBA 2.3, everything is to be case insensitive there are some problems with the PersonIdService.idl. The statement defining: "Trait trait" in TraitSelector now is illegal. This requires a minor change in the IDL to make the instance of Trait to be different. Resolution: Revised Text: Actions taken: October 13, 1999: received issue Discussion: End of Annotations:===== X-Sender: u076663@cic-mail.lanl.gov X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.0.58 Date: Wed, 13 Oct 1999 23:30:03 -0600 To: pids-rtf2@omg.org From: David Forslund Subject: Re: CASE troubles Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-UIDL: ZKjd9=1D!!]KQ!!2Z]d9 Since with CORBA 2.3, everything is to be case insensitive there are some problems with the PersonIdService.idl. The statement defining: "Trait trait" in TraitSelector now is illegal. This requires a minor change in the IDL to make the instance of Trait to be different. Dave Reply-To: From: "Martin Chapman" To: "David Forslund" , Subject: RE: CASE troubles Date: Thu, 14 Oct 1999 11:10:07 +0100 Message-ID: <001f01bf162c$4a4b8aa0$4d01020a@leo.dublin.iona.ie> MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 In-Reply-To: <4.2.0.58.19991013232634.0365bf00@cic-mail.lanl.gov> Content-Type: text/plain; charset="iso-8859-1" X-UIDL: FaG!!65Nd9eS6!!Q]_d9 This has always been illegal - some idl compilers (that shall remain nameless:-) do let them through as a "proprietary extension". Newer versions of a "nameless compiler" will enforce this restriction. AB members partcipating in PIDS should have noticed this :-) Cheers, Martin. > -----Original Message----- > From: David Forslund [mailto:dwf@lanl.gov] > Sent: 14 October 1999 06:30 > To: pids-rtf2@omg.org > Subject: Re: CASE troubles > > > Since with CORBA 2.3, everything is to be case insensitive there are > some > problems > with the PersonIdService.idl. The statement defining: "Trait > trait" in > TraitSelector now is illegal. This requires > a minor change in the IDL to make the instance of Trait to be > different. > > Dave > X-Sender: u076663@cic-mail.lanl.gov X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.0.58 Date: Thu, 14 Oct 1999 08:19:52 -0600 To: , From: David Forslund Subject: RE: CASE troubles Cc: issues@omg.org In-Reply-To: <001f01bf162c$4a4b8aa0$4d01020a@leo.dublin.iona.ie> References: <4.2.0.58.19991013232634.0365bf00@cic-mail.lanl.gov> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-UIDL: ,2O!!%G9!!<"Zd9NGP!! Unfortunately, the same problems exist in the RAD specification with the term "Operation" and the variable operation as an argument in a number of places. In a related matter TerminologyServices.idl also uses the reserved word DEFAULT when it shouldn't, which just showed up because of the case invariance enforcement. Dave At 11:10 AM 10/14/99 +0100, Martin Chapman wrote: >This has always been illegal - some idl compilers (that shall remain >nameless:-) do let them through as a "proprietary extension". >Newer versions of a "nameless compiler" will enforce this restriction. >AB members partcipating in PIDS should have noticed this :-) > >Cheers, > Martin. > > > -----Original Message----- > > From: David Forslund [mailto:dwf@lanl.gov] > > Sent: 14 October 1999 06:30 > > To: pids-rtf2@omg.org > > Subject: Re: CASE troubles > > > > > > Since with CORBA 2.3, everything is to be case insensitive there are some > > problems > > with the PersonIdService.idl. The statement defining: "Trait trait" in > > TraitSelector now is illegal. This requires > > a minor change in the IDL to make the instance of Trait to be different. > > > > Dave > > Sender: jis@fpk.hp.com Message-ID: <380601B9.F06AF433@fpk.hp.com> Date: Thu, 14 Oct 1999 12:15:53 -0400 From: Jishnu Mukerji Organization: Hewlett-Packard EIAL X-Mailer: Mozilla 4.08 [en] (X11; U; HP-UX B.11.00 9000/889) MIME-Version: 1.0 To: David Forslund CC: mchapman@iona.com, pids-rtf2@omg.org Subject: Re: CASE troubles References: <4.2.0.58.19991013232634.0365bf00@cic-mail.lanl.gov> <4.2.0.58.19991014075946.037902b0@cic-mail.lanl.gov> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-UIDL: A_f!!ATid9phD!!6$7e9 I remember I did catch the Operation problem in RAD at the AB meeting and brought it to Carol's attention. I don't remember if any of us caught that particular one in PIDS, but we did catch some. As you can see it would help a lot if nameless compilers do not incorporate brain-dead extensions.;-) In my experience, about 50% of the specs that are presented before the AB for adoption has problems of this nature (if not worse) in the IDL contained in them. Usually, I find one or two of them just browsing through the final document at the AB.:-( The FTF process certainly makes it easier to review and fix these. BTW, the case insensitive rule has been there since CORBA 0.9:-). It is nothing new in CORBA 2.3. So the "Trait trait" thing was illegal the day it was penned.:-) Jishnu David Forslund wrote: > > Unfortunately, the same problems exist in the RAD specification with > the > term "Operation" and the > variable operation as an argument in a number of places. In a > related > matter TerminologyServices.idl also uses the reserved word DEFAULT > when it > shouldn't, which just showed up because of the case invariance > enforcement. > > Dave > > At 11:10 AM 10/14/99 +0100, Martin Chapman wrote: > >This has always been illegal - some idl compilers (that shall > remain > >nameless:-) do let them through as a "proprietary extension". > >Newer versions of a "nameless compiler" will enforce this > restriction. > >AB members partcipating in PIDS should have noticed this :-) > > > >Cheers, > > Martin. > > > > > -----Original Message----- > > > From: David Forslund [mailto:dwf@lanl.gov] > > > Sent: 14 October 1999 06:30 > > > To: pids-rtf2@omg.org > > > Subject: Re: CASE troubles > > > > > > > > > Since with CORBA 2.3, everything is to be case insensitive there > are some > > > problems > > > with the PersonIdService.idl. The statement defining: "Trait > trait" in > > > TraitSelector now is illegal. This requires > > > a minor change in the IDL to make the instance of Trait to be > different. > > > > > > Dave > > > -- Jishnu Mukerji Systems Architect Email: jis@fpk.hp.com Hewlett-Packard EIAL, Tel: +1 973 443 7528 300 Campus Drive, 2E-62, Fax: +1 973 443 7422 Florham Park, NJ 07932, USA. X-Sender: carolbrt@mindspring.com X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0 Date: Thu, 14 Oct 1999 20:38:05 -0500 To: David Forslund From: Carol Burt Subject: RE: CASE troubles Cc: pids-rtf2@omg.org In-Reply-To: <4.2.0.58.19991014075946.037902b0@cic-mail.lanl.gov> References: <001f01bf162c$4a4b8aa0$4d01020a@leo.dublin.iona.ie> <4.2.0.58.19991013232634.0365bf00@cic-mail.lanl.gov> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" X-UIDL: Qah!!bo:e9&#kd92n9!! Hi, #$%#@! I just looked... the thing that is irritating is that this actually got fixed in one version of the RAD idl and then appears to have wandered back in... apparently an old version of idl got picked up and rev'd forward. I'll put it into the FTF to fix. anyway... it needs to be fixed. Carol >Unfortunately, the same problems exist in the RAD specification with the >term "Operation" and the >variable operation as an argument in a number of places. In a related >matter TerminologyServices.idl also uses the reserved word DEFAULT when it >shouldn't, which just showed up because of the case invariance enforcement. > >Dave > >At 11:10 AM 10/14/99 +0100, Martin Chapman wrote: >>This has always been illegal - some idl compilers (that shall remain >>nameless:-) do let them through as a "proprietary extension". >>Newer versions of a "nameless compiler" will enforce this restriction. >>AB members partcipating in PIDS should have noticed this :-) >> >>Cheers, >> Martin. >> >> > -----Original Message----- >> > From: David Forslund [mailto:dwf@lanl.gov] >> > Sent: 14 October 1999 06:30 >> > To: pids-rtf2@omg.org >> > Subject: Re: CASE troubles >> > >> > >> > Since with CORBA 2.3, everything is to be case insensitive there are some >> > problems >> > with the PersonIdService.idl. The statement defining: "Trait trait" in >> > TraitSelector now is illegal. This requires >> > a minor change in the IDL to make the instance of Trait to be different. >> > >> > Dave >> > > _________________________________________________________ Carol Burt 2AB, Inc. cburt@2ab.com Integration Architects 205-621-4813 www.2ab.com Member, OMG Architecture Board OMG Domain Member _________________________________________________________ "Distributed Solutions for Distributed Business (SM)" X-Sender: carolbrt@mindspring.com X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0 Date: Thu, 14 Oct 1999 20:39:59 -0500 To: Jishnu Mukerji , David Forslund From: Carol Burt Subject: Re: CASE troubles Cc: mchapman@iona.com, pids-rtf2@omg.org In-Reply-To: <380601B9.F06AF433@fpk.hp.com> References: <4.2.0.58.19991013232634.0365bf00@cic-mail.lanl.gov> <4.2.0.58.19991014075946.037902b0@cic-mail.lanl.gov> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" X-UIDL: (HO!!G@&!!X5Xd9fl*!! Hi, >I remember I did catch the Operation problem in RAD at the AB meeting and >brought it to Carol's attention. I know... it got fixed but apparently didn't make it into the irrata so the spec doesn't have the changes. The ftf will fix it. Carol I don't remember if any of us caught that >particular one in PIDS, but we did catch some. As you can see it would help a >lot if nameless compilers do not incorporate brain-dead extensions.;-) In my >experience, about 50% of the specs that are presented before the AB for adoption >has problems of this nature (if not worse) in the IDL contained in them. >Usually, I find one or two of them just browsing through the final document at >the AB.:-( The FTF process certainly makes it easier to review and fix these. > >BTW, the case insensitive rule has been there since CORBA 0.9:-). It is nothing >new in CORBA 2.3. So the "Trait trait" thing was illegal the day it was >penned.:-) > >Jishnu > >David Forslund wrote: >> >> Unfortunately, the same problems exist in the RAD specification with the >> term "Operation" and the >> variable operation as an argument in a number of places. In a related >> matter TerminologyServices.idl also uses the reserved word DEFAULT when it >> shouldn't, which just showed up because of the case invariance enforcement. >> >> Dave >> >> At 11:10 AM 10/14/99 +0100, Martin Chapman wrote: >> >This has always been illegal - some idl compilers (that shall remain >> >nameless:-) do let them through as a "proprietary extension". >> >Newer versions of a "nameless compiler" will enforce this restriction. >> >AB members partcipating in PIDS should have noticed this :-) >> > >> >Cheers, >> > Martin. >> > >> > > -----Original Message----- >> > > From: David Forslund [mailto:dwf@lanl.gov] >> > > Sent: 14 October 1999 06:30 >> > > To: pids-rtf2@omg.org >> > > Subject: Re: CASE troubles >> > > >> > > >> > > Since with CORBA 2.3, everything is to be case insensitive there are some >> > > problems >> > > with the PersonIdService.idl. The statement defining: "Trait trait" in >> > > TraitSelector now is illegal. This requires >> > > a minor change in the IDL to make the instance of Trait to be different. >> > > >> > > Dave >> > > > >-- >Jishnu Mukerji >Systems Architect > >Email: jis@fpk.hp.com Hewlett-Packard EIAL, >Tel: +1 973 443 7528 300 Campus Drive, 2E-62, >Fax: +1 973 443 7422 Florham Park, NJ 07932, USA. > _________________________________________________________ Carol Burt 2AB, Inc. cburt@2ab.com Integration Architects 205-621-4813 www.2ab.com Member, OMG Architecture Board OMG Domain Member _________________________________________________________ "Distributed Solutions for Distributed Business (SM)" X-Sender: u076663@cic-mail.lanl.gov X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.0.58 Date: Fri, 12 Nov 1999 15:15:35 -0700 To: issues@emerald.omg.org, pids-rtf2@emerald.omg.org From: David Forslund Subject: Re: issue 2937 -- PIDS RTF 2 ISSUE In-Reply-To: <4.1.19991021161027.00a1d900@emerald.omg.org> Mime-Version: 1.0 Content-Type: text/html; charset="us-ascii" X-UIDL: $>X!!IX*e9,Cid9K\)e9 At 04:11 PM 10/21/99 -0400, Juergen Boldt wrote: This is issue # 2937 CASE troubles Since with CORBA 2.3, everything is to be case insensitive there are some problems with the PersonIdService.idl. The statement defining: "Trait trait" in TraitSelector now is illegal. This requires a minor change in the IDL to make the instance of Trait to be different. formal/10-98-10-02.txt contains the following: "// In CORBA 2.3, which introduces some new keywords, new IDL lexical rules // specify that any identifier which conflicts with a new keyword // can continue to be used by pre-pending the identifier with an underscore // "_". The IDL compiler will recognize such constructs as an identifier // without the underscore. That means that all generated stubs and skeletons // use the identifier as before, maintaining compatibility for code. Of course, // the IDL must be changed, but this is a small task compared to changing all // generated code. Some of the previously defined services have some // identifiers that clash with the new keywords. To use CORBA 2.3 IDL, these // identifiers must have have the underscore pre-pended. This is called an // "Escaped Identifier". An escaped identifier is illegal in pre-CORBA 2.3." I propose that we just use this prepended identifier "_" to remove the name clash with Trait and Profile in the several places they are used. This should avoid the necessity to change any PIDS implementation code in the field. Dave