Issue 3545: org.omg.PortableInterceptor.ORBInitializerClass.XXX on command line? (interceptors-rtf) Source: International Business Machines (Mr. Russell Butek, ) Nature: Uncategorized Issue Severity: Summary: In Java, we have defined a property for ORBInitializers. But we do not define how they can be set on the command line. Do we think it is necessary? If so, then... Does the Java mapping spec provide a mechanism for setting the defined parameters via the command line? The Core provides for command line arguments of the form "-ORBXXX". The Java mapping chapter does not map "-ORBXXX" to properties. I believe it's generally assumed that "-ORBXXX" is equivalent to a property with a name of "org.omg.CORBA.XXX". If this IS defined somewhere, I would appreciate someone pointing it out to me. Now back to my question. How do we get the equivalent of "org.omg.PortableInterceptor.ORBInitializerClass.XXX" passed on the command line? Can we simply do "-ORBInitializerClass.XXX"? That would be going somewhat contrary to convention of "-ORB" meaning "org.omg.CORBA", but we could do it. Resolution: Vote YES to close this issue with no change. Issue closed Revised Text: Actions taken: April 11, 2000: received issue January 9, 2001: closed issue Discussion: End of Annotations:===== From: butek@us.ibm.com X-Lotus-FromDomain: IBMUS To: interceptors-ftf@omg.org, issues@omg.org Message-ID: <852568BE.0061BA5D.00@d54mta08.raleigh.ibm.com> Date: Tue, 11 Apr 2000 12:38:30 -0500 Subject: org.omg.PortableInterceptor.ORBInitializerClass.XXX on command line? Mime-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset=us-ascii X-UIDL: NDgd9C6P!!bn=!!&P'!! In Java, we have defined a property for ORBInitializers. But we do not define how they can be set on the command line. Do we think it is necessary? If so, then... Does the Java mapping spec provide a mechanism for setting the defined parameters via the command line? The Core provides for command line arguments of the form "-ORBXXX". The Java mapping chapter does not map "-ORBXXX" to properties. I believe it's generally assumed that "-ORBXXX" is equivalent to a property with a name of "org.omg.CORBA.XXX". If this IS defined somewhere, I would appreciate someone pointing it out to me. Now back to my question. How do we get the equivalent of "org.omg.PortableInterceptor.ORBInitializerClass.XXX" passed on the command line? Can we simply do "-ORBInitializerClass.XXX"? That would be going somewhat contrary to convention of "-ORB" meaning "org.omg.CORBA", but we could do it. Russell Butek butek@us.ibm.com Reply-To: From: "Nick Sharman" To: , Subject: RE: org.omg.PortableInterceptor.ORBInitializerClass.XXX on command line? Date: Thu, 13 Apr 2000 09:26:59 +0100 Message-ID: <005301bfa522$09561210$5610a8c0@thumper.uk.peerlogic.com> 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: <852568BE.0061BA5D.00@d54mta08.raleigh.ibm.com> Content-Type: text/plain; charset="iso-8859-1" X-UIDL: 7=(e9Rk7e95j9!!op7e9 Russell, > From: butek@us.ibm.com [mailto:butek@us.ibm.com] > > In Java, we have defined a property for ORBInitializers. But we do not > define how they can be set on the command line. Do we think it is > necessary? If so, then... > > Does the Java mapping spec provide a mechanism for setting the defined > parameters via the command line? The Core provides for command line > arguments of the form "-ORBXXX". The Java mapping chapter does not map > "-ORBXXX" to properties. I believe it's generally assumed that "-ORBXXX" > is equivalent to a property with a name of "org.omg.CORBA.XXX". > If this IS > defined somewhere, I would appreciate someone pointing it out to me. > I've never found such a statement. We support something like the scheme you mention; for example, -ORBClass maps to org.omg.CORBA.Class, but our vendor-specific ORB parameters map to com.peerlogic.broker.something. I'm hesistant about using names in the org.omg hierarchy for vendor-specific purposes. A related problem is that, if an ORB parameter gets supplied as -ORBname value on the command line and gets passed in as two consecutive members of the args array, it's not easy for a portable app to understand that "value" goes with "-ORBname", and is not one of its own arguments. The C++ and C mappings require ORB_init to shuffle argv and update argc to eliminate these values; in Java there's no modifiable equivalent of argc, tho' we can shuffle the elements of the args array and replace some of them with nulls or empty strings. There's also nothing about how they ought to be mapped to applet parameters. As applet parameters are always name/value pairs anyway, an applet can easily spot ORB parameters - but do they still need to start "-ORB", or is "ORB" good enough? (Everything above here is an issue for the IDL->Java RTF, I guess.) > Now back to my question. How do we get the equivalent of > "org.omg.PortableInterceptor.ORBInitializerClass.XXX" passed on > the command > line? Can we simply do "-ORBInitializerClass.XXX"? That would be > going > somewhat contrary to convention of "-ORB" meaning "org.omg.CORBA", > but we > could do it. > In the absence of any formal rule, I don't see why not. I wouldn't be > keen on "-ORBPortableInterceptorInitializerClass.XXX" :-) However, the XXX is really the value, but creeps into the property name because there will in general need more than one such class (but properties need unique names). How about dropping the dot, giving: "-ORBInitializerClassXXX" and "-ORBInitializerClass XXX" (the latter as one args element or two, presumably). > Russell Butek > butek@us.ibm.com > > > From: Russell_Butek@tivoli.com Received: from tivmta4.tivoli.com (tivmta4.tivoli.com [146.84.104.47]) by corp.tivoli.com (8.9.3/8.9.0) with SMTP id HAA13650; Tue, 18 Jul 2000 07:13:17 -0500 (CDT) Received: by tivmta4.tivoli.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 86256920.0043228A ; Tue, 18 Jul 2000 07:13:17 -0500 X-Lotus-FromDomain: TIVOLI SYSTEMS To: Harold Carr cc: java-rtf@omg.org, interceptors-ftf@omg.org Message-ID: <86256920.004320F3.00@tivmta4.tivoli.com> Date: Tue, 18 Jul 2000 07:13:12 -0500 Subject: Re: issue 3545 Mime-Version: 1.0 Content-Disposition: inline Content-Type: multipart/mixed; Boundary="0__=Vz3EJPNPDpDTUy7oIP5dQfxqc4lS1ImNwk0KLdpBoxf77sjne1jVPqGm" X-UIDL: H@6!!1[Ce9:gY!!~Dhd9 Harold, I really don't think we should propose anything without agreement from the Java RTF. I raised another issue there - 3643 - that is somewhat related to this one, but discussion on it has lagged. Java RTF folks, could you look at issue 3545 and tell us if you see any problems? In brief, the issue proposes mapping the Portable Interceptor property org.omg.PortableInterceptor.ORBInitializerClass as the command line argument -ORBInitializerClass. Harold Carr on 07/17/2000 01:50:16 PM To: Russell Butek/Tivoli Systems@Tivoli Systems cc: Subject: issue 3545 Russell, could you provide revised text to be voted on? Thanks, Harold [] att1.eml From: Russell_Butek@tivoli.com Received: from tivmta4.tivoli.com (tivmta4.tivoli.com [146.84.104.47]) by corp.tivoli.com (8.9.3/8.9.0) with SMTP id JAA12978; Tue, 18 Jul 2000 09:57:00 -0500 (CDT) Received: by tivmta4.tivoli.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 86256920.00521E09 ; Tue, 18 Jul 2000 09:56:56 -0500 X-Lotus-FromDomain: TIVOLI SYSTEMS To: Simon Nash cc: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Message-ID: <86256920.00521A37.00@tivmta4.tivoli.com> Date: Tue, 18 Jul 2000 09:56:43 -0500 Subject: Re: issue 3545 Mime-Version: 1.0 Content-Disposition: inline Content-Type: multipart/mixed; Boundary="0__=4GFMx2MksWiha9E030qcQv85kb5YWfoJHG40UsaVNapRnEaJcixTSE0D" X-UIDL: jID!!=RM!!%:R!!m7Zd9 I assume you mean the -D option on the java command itself? That's fine and that works. But to be complete, since ORB.init takes command line arguments, we should define the look of the command line argument equivalent to the property org.omg.PortableInterceptor.ORBInitializerClass.xxx. Simon Nash on 07/18/2000 09:37:09 AM To: Russell Butek/Tivoli Systems@Tivoli Systems cc: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 Russell, Why is the existing mechanism of defining a Java system property on the command line not adequate? Simon Russell_Butek@tivoli.com wrote: > > Harold, I really don't think we should propose anything without agreement from > the Java RTF. I raised another issue there - 3643 - that is somewhat related to > this one, but discussion on it has lagged. > > Java RTF folks, could you look at issue 3545 and tell us if you see any > problems? In brief, the issue proposes mapping the Portable Interceptor > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > argument -ORBInitializerClass. > > Harold Carr on 07/17/2000 01:50:16 PM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: > Subject: issue 3545 > > Russell, could you provide revised text to be voted on? > > Thanks, > Harold > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb [] att11.eml Date: Tue, 18 Jul 2000 15:37:09 +0100 From: Simon Nash Organization: IBM X-Mailer: Mozilla 4.72 [en] (Windows NT 5.0; I) X-Accept-Language: en MIME-Version: 1.0 To: Russell_Butek@tivoli.com CC: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 References: <86256920.004320F3.00@tivmta4.tivoli.com> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-UIDL: BEG!!XB8!!Wpb!!$D[!! Russell, Why is the existing mechanism of defining a Java system property on the command line not adequate? Simon Russell_Butek@tivoli.com wrote: > > Harold, I really don't think we should propose anything without agreement from > the Java RTF. I raised another issue there - 3643 - that is somewhat related to > this one, but discussion on it has lagged. > > Java RTF folks, could you look at issue 3545 and tell us if you see any > problems? In brief, the issue proposes mapping the Portable Interceptor > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > argument -ORBInitializerClass. > > Harold Carr on 07/17/2000 01:50:16 PM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: > Subject: issue 3545 > > Russell, could you provide revised text to be voted on? > > Thanks, > Harold > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb Date: Wed, 19 Jul 2000 10:02:51 +0100 From: Simon Nash Organization: IBM X-Mailer: Mozilla 4.72 [en] (Windows NT 5.0; I) X-Accept-Language: en MIME-Version: 1.0 To: Russell_Butek@tivoli.com CC: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 References: <86256920.00521A37.00@tivmta4.tivoli.com> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-UIDL: 1<+!!`Bod9ma;e9DcN!! Russell, ORB.init also takes a Properties object, so the ORBInitializer class can be passed in using that. This means that we already have a way to do this both on the command line and when creating an ORB instance. I don't see why anything more than this is needed for completeness. The OMG spec does not currently define command line arguments. It simply specifies that these are passed to ORB.init, which passes them on to the created ORB instance using set_parameters. So they are interpreted by the created ORB instance and represent proprietary options, not standardized ones. Simon Russell_Butek@tivoli.com wrote: > > I assume you mean the -D option on the java command itself? That's fine and > that works. But to be complete, since ORB.init takes command line arguments, we > should define the look of the command line argument equivalent to the property > org.omg.PortableInterceptor.ORBInitializerClass.xxx. > > Simon Nash on 07/18/2000 09:37:09 AM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: Harold Carr , java-rtf@omg.org, > interceptors-ftf@omg.org > Subject: Re: issue 3545 > > Russell, > Why is the existing mechanism of defining a Java system property on the > command line not adequate? > > Simon > > Russell_Butek@tivoli.com wrote: > > > > Harold, I really don't think we should propose anything without agreement from > > the Java RTF. I raised another issue there - 3643 - that is somewhat related > to > > this one, but discussion on it has lagged. > > > > Java RTF folks, could you look at issue 3545 and tell us if you see any > > problems? In brief, the issue proposes mapping the Portable Interceptor > > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > > argument -ORBInitializerClass. > > > > Harold Carr on 07/17/2000 01:50:16 PM > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > cc: > > Subject: issue 3545 > > > > Russell, could you provide revised text to be voted on? > > > > Thanks, > > Harold > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > Name: att1.eml > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > Encoding: base64 > > -- > Simon C Nash, Technology Architect, IBM Java Technology Centre > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb From: Russell_Butek@tivoli.com Received: from tivmta4.tivoli.com (tivmta4.tivoli.com [146.84.104.47]) by corp.tivoli.com (8.9.3/8.9.0) with SMTP id HAA13145; Wed, 19 Jul 2000 07:18:16 -0500 (CDT) Received: by tivmta4.tivoli.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 86256921.004398F8 ; Wed, 19 Jul 2000 07:18:20 -0500 X-Lotus-FromDomain: TIVOLI SYSTEMS To: Simon Nash cc: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Message-ID: <86256921.004397E5.00@tivmta4.tivoli.com> Date: Wed, 19 Jul 2000 07:18:14 -0500 Subject: Re: issue 3545 Mime-Version: 1.0 Content-Disposition: inline Content-Type: multipart/mixed; Boundary="0__=sjiT8saKCTGn03ACMYCyV5wOcQZyNzCn6Xea5AyizFhUErQoBNoEgSWs" X-UIDL: `a3!!-RJe9b7K!!DfOe9 I am not saying we don't have a way to do this. I'm saying that there is a hole in the spec. The core spec defines a number of "-ORBxxx" strings that come into ORB_init via the arg_list parameter. The Java mapping for ORB.init has an args parameter which is, presumably, equivalent to the arg_list parameter defined in the core. The Java mapping also provides a Properties object as a parameter. It then specifies a couple valid properties, but it does not define how the equivalent can be defined via the args parameter. Now that we're getting more properties we should provide a standard way of mapping properties to args and vice versa. Or are you saying that, for Java, there is no standard usage for the args parameter. That would be strange since the core itself defines a usage for the arg_list parameter. Simon Nash on 07/19/2000 04:02:51 AM To: Russell Butek/Tivoli Systems@Tivoli Systems cc: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 Russell, ORB.init also takes a Properties object, so the ORBInitializer class can be passed in using that. This means that we already have a way to do this both on the command line and when creating an ORB instance. I don't see why anything more than this is needed for completeness. The OMG spec does not currently define command line arguments. It simply specifies that these are passed to ORB.init, which passes them on to the created ORB instance using set_parameters. So they are interpreted by the created ORB instance and represent proprietary options, not standardized ones. Simon Russell_Butek@tivoli.com wrote: > > I assume you mean the -D option on the java command itself? That's fine and > that works. But to be complete, since ORB.init takes command line arguments, we > should define the look of the command line argument equivalent to the property > org.omg.PortableInterceptor.ORBInitializerClass.xxx. > > Simon Nash on 07/18/2000 09:37:09 AM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: Harold Carr , java-rtf@omg.org, > interceptors-ftf@omg.org > Subject: Re: issue 3545 > > Russell, > Why is the existing mechanism of defining a Java system property on the > command line not adequate? > > Simon > > Russell_Butek@tivoli.com wrote: > > > > Harold, I really don't think we should propose anything without agreement from > > the Java RTF. I raised another issue there - 3643 - that is somewhat related > to > > this one, but discussion on it has lagged. > > > > Java RTF folks, could you look at issue 3545 and tell us if you see any > > problems? In brief, the issue proposes mapping the Portable Interceptor > > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > > argument -ORBInitializerClass. > > > > Harold Carr on 07/17/2000 01:50:16 PM > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > cc: > > Subject: issue 3545 > > > > Russell, could you provide revised text to be voted on? > > > > Thanks, > > Harold > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > Name: att1.eml > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > Encoding: base64 > > -- > Simon C Nash, Technology Architect, IBM Java Technology Centre > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb [] att12.eml Date: Wed, 19 Jul 2000 10:02:51 +0100 From: Simon Nash Organization: IBM X-Mailer: Mozilla 4.72 [en] (Windows NT 5.0; I) X-Accept-Language: en MIME-Version: 1.0 To: Russell_Butek@tivoli.com CC: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 References: <86256920.00521A37.00@tivmta4.tivoli.com> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-UIDL: 1<+!!`Bod9ma;e9DcN!! Russell, ORB.init also takes a Properties object, so the ORBInitializer class can be passed in using that. This means that we already have a way to do this both on the command line and when creating an ORB instance. I don't see why anything more than this is needed for completeness. The OMG spec does not currently define command line arguments. It simply specifies that these are passed to ORB.init, which passes them on to the created ORB instance using set_parameters. So they are interpreted by the created ORB instance and represent proprietary options, not standardized ones. Simon Russell_Butek@tivoli.com wrote: > > I assume you mean the -D option on the java command itself? That's fine and > that works. But to be complete, since ORB.init takes command line arguments, we > should define the look of the command line argument equivalent to the property > org.omg.PortableInterceptor.ORBInitializerClass.xxx. > > Simon Nash on 07/18/2000 09:37:09 AM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: Harold Carr , java-rtf@omg.org, > interceptors-ftf@omg.org > Subject: Re: issue 3545 > > Russell, > Why is the existing mechanism of defining a Java system property on the > command line not adequate? > > Simon > > Russell_Butek@tivoli.com wrote: > > > > Harold, I really don't think we should propose anything without agreement from > > the Java RTF. I raised another issue there - 3643 - that is somewhat related > to > > this one, but discussion on it has lagged. > > > > Java RTF folks, could you look at issue 3545 and tell us if you see any > > problems? In brief, the issue proposes mapping the Portable Interceptor > > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > > argument -ORBInitializerClass. > > > > Harold Carr on 07/17/2000 01:50:16 PM > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > cc: > > Subject: issue 3545 > > > > Russell, could you provide revised text to be voted on? > > > > Thanks, > > Harold > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > Name: att1.eml > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > Encoding: base64 > > -- > Simon C Nash, Technology Architect, IBM Java Technology Centre > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb From: Russell_Butek@tivoli.com Received: from tivmta4.tivoli.com (tivmta4.tivoli.com [146.84.104.47]) by corp.tivoli.com (8.9.3/8.9.0) with SMTP id HAA13145; Wed, 19 Jul 2000 07:18:16 -0500 (CDT) Received: by tivmta4.tivoli.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 86256921.004398F8 ; Wed, 19 Jul 2000 07:18:20 -0500 X-Lotus-FromDomain: TIVOLI SYSTEMS To: Simon Nash cc: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Message-ID: <86256921.004397E5.00@tivmta4.tivoli.com> Date: Wed, 19 Jul 2000 07:18:14 -0500 Subject: Re: issue 3545 Mime-Version: 1.0 Content-Disposition: inline Content-Type: multipart/mixed; Boundary="0__=sjiT8saKCTGn03ACMYCyV5wOcQZyNzCn6Xea5AyizFhUErQoBNoEgSWs" X-UIDL: `a3!!-RJe9b7K!!DfOe9 I am not saying we don't have a way to do this. I'm saying that there is a hole in the spec. The core spec defines a number of "-ORBxxx" strings that come into ORB_init via the arg_list parameter. The Java mapping for ORB.init has an args parameter which is, presumably, equivalent to the arg_list parameter defined in the core. The Java mapping also provides a Properties object as a parameter. It then specifies a couple valid properties, but it does not define how the equivalent can be defined via the args parameter. Now that we're getting more properties we should provide a standard way of mapping properties to args and vice versa. Or are you saying that, for Java, there is no standard usage for the args parameter. That would be strange since the core itself defines a usage for the arg_list parameter. Simon Nash on 07/19/2000 04:02:51 AM To: Russell Butek/Tivoli Systems@Tivoli Systems cc: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 Russell, ORB.init also takes a Properties object, so the ORBInitializer class can be passed in using that. This means that we already have a way to do this both on the command line and when creating an ORB instance. I don't see why anything more than this is needed for completeness. The OMG spec does not currently define command line arguments. It simply specifies that these are passed to ORB.init, which passes them on to the created ORB instance using set_parameters. So they are interpreted by the created ORB instance and represent proprietary options, not standardized ones. Simon Russell_Butek@tivoli.com wrote: > > I assume you mean the -D option on the java command itself? That's fine and > that works. But to be complete, since ORB.init takes command line arguments, we > should define the look of the command line argument equivalent to the property > org.omg.PortableInterceptor.ORBInitializerClass.xxx. > > Simon Nash on 07/18/2000 09:37:09 AM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: Harold Carr , java-rtf@omg.org, > interceptors-ftf@omg.org > Subject: Re: issue 3545 > > Russell, > Why is the existing mechanism of defining a Java system property on the > command line not adequate? > > Simon > > Russell_Butek@tivoli.com wrote: > > > > Harold, I really don't think we should propose anything without agreement from > > the Java RTF. I raised another issue there - 3643 - that is somewhat related > to > > this one, but discussion on it has lagged. > > > > Java RTF folks, could you look at issue 3545 and tell us if you see any > > problems? In brief, the issue proposes mapping the Portable Interceptor > > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > > argument -ORBInitializerClass. > > > > Harold Carr on 07/17/2000 01:50:16 PM > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > cc: > > Subject: issue 3545 > > > > Russell, could you provide revised text to be voted on? > > > > Thanks, > > Harold > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > Name: att1.eml > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > Encoding: base64 > > -- > Simon C Nash, Technology Architect, IBM Java Technology Centre > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb [] att12.eml Date: Wed, 19 Jul 2000 15:32:13 +0100 From: Simon Nash Organization: IBM X-Mailer: Mozilla 4.72 [en] (Windows NT 5.0; I) X-Accept-Language: en MIME-Version: 1.0 To: Russell_Butek@tivoli.com CC: Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 References: <86256921.004397E5.00@tivmta4.tivoli.com> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-UIDL: M#kd9!(""!F%C!!0b#!! Russell, Yes, I am saying that in Java there are no standard values defined for -ORBxxx options in the args parameter. In the 2.3 core spec, the only standard value defined is -ORBid and Java does not support this functionality. So I do not think this is inconsistent. Did the PI spec extend the list of standard arguments for the core PIDL ORB_init? If so, then we may need to re-examine the Java situation in the light of this. However, since Java's ORB.init takes a properties argument and the core PIDL ORB_init does not, it would still be valid according to the rules of PIDL to say that the PIDL ORB_init argument strings map into Java properties on ORB.init. Simon Russell_Butek@tivoli.com wrote: > > I am not saying we don't have a way to do this. I'm saying that there is a hole > in the spec. The core spec defines a number of "-ORBxxx" strings that come into > ORB_init via the arg_list parameter. The Java mapping for ORB.init has an args > parameter which is, presumably, equivalent to the arg_list parameter defined in > the core. The Java mapping also provides a Properties object as a parameter. > It then specifies a couple valid properties, but it does not define how the > equivalent can be defined via the args parameter. Now that we're getting more > properties we should provide a standard way of mapping properties to args and > vice versa. > > Or are you saying that, for Java, there is no standard usage for the args > parameter. That would be strange since the core itself defines a usage for the > arg_list parameter. > > Simon Nash on 07/19/2000 04:02:51 AM > > To: Russell Butek/Tivoli Systems@Tivoli Systems > cc: Harold Carr , java-rtf@omg.org, > interceptors-ftf@omg.org > Subject: Re: issue 3545 > > Russell, > ORB.init also takes a Properties object, so the ORBInitializer class can > be passed in using that. This means that we already have a way to do this > both on the command line and when creating an ORB instance. I don't see why > anything more than this is needed for completeness. > > The OMG spec does not currently define command line arguments. It simply > specifies that these are passed to ORB.init, which passes them on to the created > ORB instance using set_parameters. So they are interpreted by the created ORB > instance and represent proprietary options, not standardized ones. > > Simon > > Russell_Butek@tivoli.com wrote: > > > > I assume you mean the -D option on the java command itself? That's fine and > > that works. But to be complete, since ORB.init takes command line arguments, > we > > should define the look of the command line argument equivalent to the property > > org.omg.PortableInterceptor.ORBInitializerClass.xxx. > > > > Simon Nash on 07/18/2000 09:37:09 AM > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > cc: Harold Carr , java-rtf@omg.org, > > interceptors-ftf@omg.org > > Subject: Re: issue 3545 > > > > Russell, > > Why is the existing mechanism of defining a Java system property on the > > command line not adequate? > > > > Simon > > > > Russell_Butek@tivoli.com wrote: > > > > > > Harold, I really don't think we should propose anything without agreement > from > > > the Java RTF. I raised another issue there - 3643 - that is somewhat > related > > to > > > this one, but discussion on it has lagged. > > > > > > Java RTF folks, could you look at issue 3545 and tell us if you see any > > > problems? In brief, the issue proposes mapping the Portable Interceptor > > > property org.omg.PortableInterceptor.ORBInitializerClass as the command line > > > argument -ORBInitializerClass. > > > > > > Harold Carr on 07/17/2000 01:50:16 PM > > > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > > cc: > > > Subject: issue 3545 > > > > > > Russell, could you provide revised text to be voted on? > > > > > > Thanks, > > > Harold > > > > > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > > > > Name: att1.eml > > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > > Encoding: base64 > > > > -- > > Simon C Nash, Technology Architect, IBM Java Technology Centre > > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > Name: att1.eml > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > Encoding: base64 > > -- > Simon C Nash, Technology Architect, IBM Java Technology Centre > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > Name: att1.eml > att1.eml Type: Internet E-Mail Message (message/rfc822) > Encoding: base64 -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb Date: Wed, 19 Jul 2000 17:08:06 +0100 From: Simon Nash Organization: IBM X-Mailer: Mozilla 4.72 [en] (Windows NT 5.0; I) X-Accept-Language: en MIME-Version: 1.0 To: Mark Spruiell CC: Mary Leland , juergen@omg.org Subject: Re: issue 3545 References: <86256921.004397E5.00@tivmta4.tivoli.com> <4.3.2.7.2.20000719083116.0218d100@192.168.1.1> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-UIDL: WcD!!!Q+!!_)6e92,Je9 Mark, This is a bug in the spec. I raised an issue about this in 3/99. Here's the email I sent to the OMG issues list: > Date: Mon, 08 Mar 1999 12:30:43 +0000 > From: Simon Nash > Organization: IBM > X-Mailer: Mozilla 4.5 [en] (Win98; I) > X-Accept-Language: en > MIME-Version: 1.0 > To: issues@omg.org > CC: java-rtf@omg.org > Subject: Incorrect search order for ORB implementation class name > Content-Type: text/plain; charset=us-ascii > Content-Transfer-Encoding: 7bit > > The first bullet of the search order for an ORB implementation class > name in section 25.21.7 is: > . check in Applet parameter or application string array, if any > > This seems wrong. The spec does not define a standard way to > provide > an ORB implementation class name in the application string array. > The properties parameter that is also passed to ORB.init does allow > the ORB implementation class to be set, so it seems redundant to > also > allow it to be set using the application array. > > Either the spec should state how the implementation class name is > passed in the application string array, or the first bullet should > be changed to: > . check in Applet parameter, if any Unfortunately this email was merged in error with another issue that I submitted at the same time and this now shows up on the OMG issues database as issue 2518, the text of which is mangled beyond recognition and understanding. Mary, can you get this fixed so we can put this up for a vote? I believe the correct resolution is to change the first bullet to disallow the implementation class name from being passed using the application array. But this could depend on how issue 3545 is resolved. Simon Mark Spruiell wrote: > > Hi Simon, > > I didn't post this to the mailing lists because it's not directly > related to the discussion at hand. > > However, I'm curious about something, and was wondering if you could > shed any light on it. The IDL/Java mapping in section 1.21.9 says > > "In any event, when creating an ORB instance, the class names of > the ORB implementation are located using the following search order: > > * check in Applet parameter or application string array, if any > ..." > > I'm assuming here that the "application string array" is the > String[] argument passed to ORB.init(). > > My question is: How are the ORB implementation class names supposed > to be specified in this string array? > > Perhaps like this? > > String[] args = { "org.omg.CORBA.ORBClass", "com.vendor.orbimpl" } > > Take care, > > Mark > -- > Mark E. Spruiell > Object Oriented Concepts, Inc. > mes@ooc.com * http://www.ooc.com * 1-978-439-9285 x 247 > > At 07:32 AM 7/19/00, you wrote: > >Russell, > >Yes, I am saying that in Java there are no standard values defined for -ORBxxx > >options in the args parameter. In the 2.3 core spec, the only standard value > >defined is -ORBid and Java does not support this functionality. So I do not > >think this is inconsistent. > > > >Did the PI spec extend the list of standard arguments for the core PIDL > >ORB_init? > >If so, then we may need to re-examine the Java situation in the light of this. > >However, since Java's ORB.init takes a properties argument and the core PIDL > >ORB_init does not, it would still be valid according to the rules of PIDL > >to say > >that the PIDL ORB_init argument strings map into Java properties on ORB.init. > > > > Simon > > > >Russell_Butek@tivoli.com wrote: > > > > > > I am not saying we don't have a way to do this. I'm saying that there > > is a hole > > > in the spec. The core spec defines a number of "-ORBxxx" strings that > > come into > > > ORB_init via the arg_list parameter. The Java mapping for ORB.init has > > an args > > > parameter which is, presumably, equivalent to the arg_list parameter > > defined in > > > the core. The Java mapping also provides a Properties object as a > > parameter. > > > It then specifies a couple valid properties, but it does not define how the > > > equivalent can be defined via the args parameter. Now that we're > > getting more > > > properties we should provide a standard way of mapping properties to > > args and > > > vice versa. > > > > > > Or are you saying that, for Java, there is no standard usage for the args > > > parameter. That would be strange since the core itself defines a usage > > for the > > > arg_list parameter. > > > > > > Simon Nash on 07/19/2000 04:02:51 AM > > > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > > cc: Harold Carr , java-rtf@omg.org, > > > interceptors-ftf@omg.org > > > Subject: Re: issue 3545 > > > > > > Russell, > > > ORB.init also takes a Properties object, so the ORBInitializer class can > > > be passed in using that. This means that we already have a way to do this > > > both on the command line and when creating an ORB instance. I don't > > see why > > > anything more than this is needed for completeness. > > > > > > The OMG spec does not currently define command line arguments. It simply > > > specifies that these are passed to ORB.init, which passes them on to > > the created > > > ORB instance using set_parameters. So they are interpreted by the > > created ORB > > > instance and represent proprietary options, not standardized ones. > > > > > > Simon > > > > > > Russell_Butek@tivoli.com wrote: > > > > > > > > I assume you mean the -D option on the java command itself? That's > > fine and > > > > that works. But to be complete, since ORB.init takes command line > > arguments, > > > we > > > > should define the look of the command line argument equivalent to the > > property > > > > org.omg.PortableInterceptor.ORBInitializerClass.xxx. > > > > > > > > Simon Nash on 07/18/2000 09:37:09 AM > > > > > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > > > cc: Harold Carr , java-rtf@omg.org, > > > > interceptors-ftf@omg.org > > > > Subject: Re: issue 3545 > > > > > > > > Russell, > > > > Why is the existing mechanism of defining a Java system property on the > > > > command line not adequate? > > > > > > > > Simon > > > > > > > > Russell_Butek@tivoli.com wrote: > > > > > > > > > > Harold, I really don't think we should propose anything without > > agreement > > > from > > > > > the Java RTF. I raised another issue there - 3643 - that is somewhat > > > related > > > > to > > > > > this one, but discussion on it has lagged. > > > > > > > > > > Java RTF folks, could you look at issue 3545 and tell us if you see any > > > > > problems? In brief, the issue proposes mapping the Portable > > Interceptor > > > > > property org.omg.PortableInterceptor.ORBInitializerClass as the > > command line > > > > > argument -ORBInitializerClass. > > > > > > > > > > Harold Carr on 07/17/2000 01:50:16 PM > > > > > > > > > > To: Russell Butek/Tivoli Systems@Tivoli Systems > > > > > cc: > > > > > Subject: issue 3545 > > > > > > > > > > Russell, could you provide revised text to be voted on? > > > > > > > > > > Thanks, > > > > > Harold > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > > > > > > > > > > Name: att1.eml > > > > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > > > > Encoding: base64 > > > > > > > > -- > > > > Simon C Nash, Technology Architect, IBM Java Technology Centre > > > > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > > > > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > > > > > > > > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > > > > > Name: att1.eml > > > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > > > Encoding: base64 > > > > > > -- > > > Simon C Nash, Technology Architect, IBM Java Technology Centre > > > Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > > > Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb > > > > > > > > --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > Name: att1.eml > > > att1.eml Type: Internet E-Mail Message (message/rfc822) > > > Encoding: base64 > > > >-- > >Simon C Nash, Technology Architect, IBM Java Technology Centre > >Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England > >Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb -- Simon C Nash, Technology Architect, IBM Java Technology Centre Tel. +44-1962-815156 Fax +44-1962-818999 Hursley, England Internet: nash@hursley.ibm.com Lotus Notes: Simon Nash@ibmgb From: Paul Kyzivat To: java-rtf@omg.org, interceptors-ftf@omg.org Subject: RE: issue 3545 Date: Wed, 19 Jul 2000 13:14:02 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Content-Type: text/plain; charset="iso-8859-1" X-UIDL: e?\d9J4~e9=9`!!5gM!! > From: Simon Nash [mailto:nash@hursley.ibm.com] > Russell, > Yes, I am saying that in Java there are no standard values > defined for -ORBxxx > options in the args parameter. In the 2.3 core spec, the > only standard value > defined is -ORBid and Java does not support this > functionality. So I do not > think this is inconsistent. The INS spec also mandated some -ORBxxx values. Date: Thu, 20 Jul 2000 05:16:55 +1000 (EST) From: Michi Henning To: Simon Nash cc: Russell_Butek@tivoli.com, Harold Carr , java-rtf@omg.org, interceptors-ftf@omg.org Subject: Re: issue 3545 In-Reply-To: <3975BBED.F7E6F244@hursley.ibm.com> Message-ID: Organization: Object Oriented Concepts MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-UIDL: ~$Sd9oVTd9@T#e9ipmd9 On Wed, 19 Jul 2000, Simon Nash wrote: > Russell, > Yes, I am saying that in Java there are no standard values defined > for -ORBxxx > options in the args parameter. In the 2.3 core spec, the only > standard value > defined is -ORBid and Java does not support this functionality. So > I do not > think this is inconsistent. > > Did the PI spec extend the list of standard arguments for the core > PIDL ORB_init? > If so, then we may need to re-examine the Java situation in the > light of this. For 2.4, we have the INS arguments, -ORBInitRef and -ORBDefaultInitRef. Cheers, Michi. -- Michi Henning +61 7 3891 5744 Object Oriented Concepts +61 4 1118 2700 (mobile) Suite 4, 904 Stanley St +61 7 3891 5009 (fax) East Brisbane 4169 michi@ooc.com.au AUSTRALIA http://www.ooc.com.au/staff/michi-henning.html From: Russell_Butek@tivoli.com Received: from tivmta4.tivoli.com (tivmta4.tivoli.com [146.84.104.47]) by corp.tivoli.com (8.9.3/8.9.0) with SMTP id QAA23932; Mon, 31 Jul 2000 16:48:52 -0500 (CDT) Received: by tivmta4.tivoli.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 8625692D.00779E65 ; Mon, 31 Jul 2000 16:46:33 -0500 X-Lotus-FromDomain: TIVOLI SYSTEMS To: Harold Carr cc: interceptors-ftf@omg.org Message-ID: <8625692D.007779AF.00@tivmta4.tivoli.com> Date: Mon, 31 Jul 2000 16:44:56 -0500 Subject: Re: Final vote issues due on Wednesday Mime-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset=us-ascii X-UIDL: O=\!!9S5e9~-]!!F"o!! Put 3545 on the next vote as "Close, no change". I misunderstood the relationship in Java between the properties object and the args object on ORB.init when I raised this issue. There apparently isn't any relationship and I thought there was. Russell Butek rbutek@tivoli.com Harold Carr on 07/31/2000 04:04:08 PM To: interceptors-ftf@omg.org cc: (bcc: Russell Butek/Tivoli Systems) Subject: Final vote issues due on Wednesday Our next vote, vote 7, begins on Thursday August 3. This is the LAST VOTE in this FTF. The deadline for submitting proposals for voting is Wednesday August 2 (to give me time to put the vote document together). Here are issues that have not been included in a vote up to this point: 3322 - ORBInitInfo::object_to_string 3429 - ORB/PIDL 3435 - Interceptors and finalization (Matthew and Michi please send the proposal you want voted on.) 3545 - -ORBInitializer command line 3599 - detail when request interceptors called 3609 - Overriding POA policies 3615 - ReceiveIORInterceptor (Nick sent out a proposal with questions, but no one responded to his questions. Should we just include the proposal?) 3693 - byte order in encoding I will be writing proposals for: 3625 - resolve in pre_init 3677 - security - invocation id 3678 - security - Retry exception 3679 - security - invocation begin/end interceptors Cheers, Harold