odd PH/VMS logical name behaviour

Hamilton, Allison Allison.Hamilton@Cognos.COM
Wed, 10 Mar 2004 15:36:05 -0500


This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C406DF.4F59FC2C
Content-Type: text/plain

Chris - I agree, OpenVMS has always been upwardly compatible.  However, I
have noticed that with the POSIX stuff, this is no longer always the case.
When I did the investigation for ODS-5 support to see if we could support
any of it with the existing PowerHouse code I had a long interaction with
Compaq and was told that there was no longer any way to parse a file name
based on ODS-2 rules using the name parsing routines.  There's no switch
that you can use (unless they've added it and it didn't sound like they were
planning to) to say please parse this file name based on ODS-2 rules. It
isn't until you actually try to go create the file on the ODS-2 disc that it
says - sorry, can't do that here.  That, is an upward non-compatibility as
far as I'm concerned, you can't validate files conclusively any more unless
you are on an ODS-5 disc.

We use the default options in sys$trnlnm calls, whatever the CASE_BLIND
setting is default is what we use.  We do, however, upshift the logical name
before calling sys$trnlnm.  

I think that the new test cluster that our systems people are currently
setting up will be 7.3 OpenVMS.  I will give it a try when it is up and
running if it is 7.3 and see what I find.

Allison Hamilton.

-----Original Message-----
From: powerh-l-admin@lists.sowder.com
[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Chris Sharman
Sent: Wednesday, March 10, 2004 12:11 PM
To: Hamilton, Allison
Cc: powerh-l@lists.sowder.com; doug.moffatt@cognos.com
Subject: Re: odd PH/VMS logical name behaviour


Hamilton, Allison wrote:
> Chris,  I think this may be a difference in behaviour between 7.2 and 
> 7.3 of OpenVMS because I tried this in 7.10G3, 7.10G5 and 8.30D4 (all on 
> 6.2 of OpenVMS) and 8.40D on 7.2-1 and 8.40D on 7.2-2 of OpenVMS and 
> they retrieve the value of Test correctly.  I don't have a 7.3 OpenVMS 
> system available to me yet, so I can't test it there.  Perhaps there is 
> a setting on the logical or in your environment that would affect the 
> default behaviour of the logical translation function (trnlnm) that is 
> available in 7.3???
> 
> I know that OpenVMS is moving more and more into the Posix world which 
> requires case sensitivity, so more things behave somewhat differently.  
> I don't know if this one is an intentional change or not on the part of 
> OpenVMS.  Usually they don't change the default behaviour of things so 
> drastically, but who knows.

Curious. Nothing I'm aware of - DCL has always uppercased everything, 
and still does, unless you enable the mixed-case behaviour which goes 
with the ODS5 file system, which we haven't.

> Prior to 7.3 retrieval using uppercase always retrieved the correct 
> value due to the basically uppercase or case insensitive nature of the 
> VMS environment.  PowerHouse is built with this assumption currently, 
> and generally upcases just about everything coming in to ensure it can 
> find it.

sys$trnlnm (and lib$get_logical) have always had a LNM$M_CASE_BLIND 
option, but it has never been the default behaviour, and still isn't.
DCL f$trnlnm offers case-blind by default, and case-sensitive 
optionally. The older (obsolete since about V5) sys$trnlog and 
associated lib$ routine may do case-blind by default - I don't recall.

I don't think VMS has ever seen a non-compatible system service change 
like you suggest.
Anyone got access to VMS 7.3 and an earlier version to confirm/deny ?

> Allison Hamilton
> Senior Software Engineer
> Cognos Inc.
> Any opinions expressed within this message are mine alone and my 
> employer cannot be held accountable for them  :-}
> 
> -----Original Message-----
> From: powerh-l-admin@lists.sowder.com
> [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Chris Sharman
> Sent: Wednesday, March 10, 2004 6:30 AM
> To: PowerH List
> Subject: odd PH/VMS logical name behaviour
> 
> 
> I've got a mixed-case logical name:
> $ def "Test" "MyTest"
> 
> $ quiz
> Qz> def mytest char*10 = getsystemval("Test")
> Qz> rep mytest
> 
> Reports it as blank - presumably it doesn't find it.
> Defining it as uppercase works fine.
> 
> VMS logical names can be defined mixed-case (but are usually uppercase).
> They can be fetched either case-sensitive or case-blind.
> 
> Powerhouse would appear to be uppercasing the supplied string, then
> doing a case-sensitive search - not the optimum behaviour !
> Be nice if it used the string I gave it, and probably set the case-blind
> flag to avoid breaking most existing stuff.
> 
> VMS 7.3, PH 7.10G3


-----------------------------------------------------------------------

Any views expressed in this message are those of the sender and not
necessarily those of CCA Group.  The unauthorized use, disclosure,
copying or alteration of this message is forbidden.  The contents of
this message may be confidential and/or privileged, copyright CCA Group
and are intended solely for the use of the individual or entity to whom
they are addressed.  Whilst this message has been scanned, CCA Group
cannot guarantee that it is virus free or compatible with your systems
and accepts no responsibility for any loss or damage arising from its
use. The recipient is advised to run their own anti-virus software. If
you receive this message in error please contact
postmaster@ccagroup.co.uk immediately, destroy any copies and delete it
from your computer systems.
= = = = = = = = = = = = = = = = = = = = = = = = = = = =
Mailing list: powerh-l@lists.sowder.com
Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com
Unsubscribe: "unsubscribe <password>" in message body to
powerh-l-request@lists.sowder.com
http://lists.sowder.com/mailman/listinfo/powerh-l
This list is closed, thus to post to the list you must be a subscriber.

This message may contain privileged and/or confidential information.  If you
have received this e-mail in error or are not the intended recipient, you
may not use, copy, disseminate or distribute it; do not open any
attachments, delete it immediately from your system and notify the sender
promptly by e-mail that you have done so.  Thank you.

------_=_NextPart_001_01C406DF.4F59FC2C
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3DUS-ASCII">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2657.73">
<TITLE>RE: odd PH/VMS logical name behaviour</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Chris - I agree, OpenVMS has always been upwardly =
compatible.&nbsp; However, I have noticed that with the POSIX stuff, =
this is no longer always the case.&nbsp; When I did the investigation =
for ODS-5 support to see if we could support any of it with the =
existing PowerHouse code I had a long interaction with Compaq and was =
told that there was no longer any way to parse a file name based on =
ODS-2 rules using the name parsing routines.&nbsp; There's no switch =
that you can use (unless they've added it and it didn't sound like they =
were planning to) to say please parse this file name based on ODS-2 =
rules. It isn't until you actually try to go create the file on the =
ODS-2 disc that it says - sorry, can't do that here.&nbsp; That, is an =
upward non-compatibility as far as I'm concerned, you can't validate =
files conclusively any more unless you are on an ODS-5 disc.</FONT></P>

<P><FONT SIZE=3D2>We use the default options in sys$trnlnm calls, =
whatever the CASE_BLIND setting is default is what we use.&nbsp; We do, =
however, upshift the logical name before calling sys$trnlnm.&nbsp; =
</FONT></P>

<P><FONT SIZE=3D2>I think that the new test cluster that our systems =
people are currently setting up will be 7.3 OpenVMS.&nbsp; I will give =
it a try when it is up and running if it is 7.3 and see what I =
find.</FONT></P>

<P><FONT SIZE=3D2>Allison Hamilton.</FONT>
</P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: powerh-l-admin@lists.sowder.com</FONT>
<BR><FONT SIZE=3D2>[<A =
HREF=3D"mailto:powerh-l-admin@lists.sowder.com">mailto:powerh-l-admin@li=
sts.sowder.com</A>]On Behalf Of Chris Sharman</FONT>
<BR><FONT SIZE=3D2>Sent: Wednesday, March 10, 2004 12:11 PM</FONT>
<BR><FONT SIZE=3D2>To: Hamilton, Allison</FONT>
<BR><FONT SIZE=3D2>Cc: powerh-l@lists.sowder.com; =
doug.moffatt@cognos.com</FONT>
<BR><FONT SIZE=3D2>Subject: Re: odd PH/VMS logical name =
behaviour</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>Hamilton, Allison wrote:</FONT>
<BR><FONT SIZE=3D2>&gt; Chris,&nbsp; I think this may be a difference =
in behaviour between 7.2 and </FONT>
<BR><FONT SIZE=3D2>&gt; 7.3 of OpenVMS because I tried this in 7.10G3, =
7.10G5 and 8.30D4 (all on </FONT>
<BR><FONT SIZE=3D2>&gt; 6.2 of OpenVMS) and 8.40D on 7.2-1 and 8.40D on =
7.2-2 of OpenVMS and </FONT>
<BR><FONT SIZE=3D2>&gt; they retrieve the value of Test =
correctly.&nbsp; I don't have a 7.3 OpenVMS </FONT>
<BR><FONT SIZE=3D2>&gt; system available to me yet, so I can't test it =
there.&nbsp; Perhaps there is </FONT>
<BR><FONT SIZE=3D2>&gt; a setting on the logical or in your environment =
that would affect the </FONT>
<BR><FONT SIZE=3D2>&gt; default behaviour of the logical translation =
function (trnlnm) that is </FONT>
<BR><FONT SIZE=3D2>&gt; available in 7.3???</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; I know that OpenVMS is moving more and more =
into the Posix world which </FONT>
<BR><FONT SIZE=3D2>&gt; requires case sensitivity, so more things =
behave somewhat differently.&nbsp; </FONT>
<BR><FONT SIZE=3D2>&gt; I don't know if this one is an intentional =
change or not on the part of </FONT>
<BR><FONT SIZE=3D2>&gt; OpenVMS.&nbsp; Usually they don't change the =
default behaviour of things so </FONT>
<BR><FONT SIZE=3D2>&gt; drastically, but who knows.</FONT>
</P>

<P><FONT SIZE=3D2>Curious. Nothing I'm aware of - DCL has always =
uppercased everything, </FONT>
<BR><FONT SIZE=3D2>and still does, unless you enable the mixed-case =
behaviour which goes </FONT>
<BR><FONT SIZE=3D2>with the ODS5 file system, which we haven't.</FONT>
</P>

<P><FONT SIZE=3D2>&gt; Prior to 7.3 retrieval using uppercase always =
retrieved the correct </FONT>
<BR><FONT SIZE=3D2>&gt; value due to the basically uppercase or case =
insensitive nature of the </FONT>
<BR><FONT SIZE=3D2>&gt; VMS environment.&nbsp; PowerHouse is built with =
this assumption currently, </FONT>
<BR><FONT SIZE=3D2>&gt; and generally upcases just about everything =
coming in to ensure it can </FONT>
<BR><FONT SIZE=3D2>&gt; find it.</FONT>
</P>

<P><FONT SIZE=3D2>sys$trnlnm (and lib$get_logical) have always had a =
LNM$M_CASE_BLIND </FONT>
<BR><FONT SIZE=3D2>option, but it has never been the default behaviour, =
and still isn't.</FONT>
<BR><FONT SIZE=3D2>DCL f$trnlnm offers case-blind by default, and =
case-sensitive </FONT>
<BR><FONT SIZE=3D2>optionally. The older (obsolete since about V5) =
sys$trnlog and </FONT>
<BR><FONT SIZE=3D2>associated lib$ routine may do case-blind by default =
- I don't recall.</FONT>
</P>

<P><FONT SIZE=3D2>I don't think VMS has ever seen a non-compatible =
system service change </FONT>
<BR><FONT SIZE=3D2>like you suggest.</FONT>
<BR><FONT SIZE=3D2>Anyone got access to VMS 7.3 and an earlier version =
to confirm/deny ?</FONT>
</P>

<P><FONT SIZE=3D2>&gt; Allison Hamilton</FONT>
<BR><FONT SIZE=3D2>&gt; Senior Software Engineer</FONT>
<BR><FONT SIZE=3D2>&gt; Cognos Inc.</FONT>
<BR><FONT SIZE=3D2>&gt; Any opinions expressed within this message are =
mine alone and my </FONT>
<BR><FONT SIZE=3D2>&gt; employer cannot be held accountable for =
them&nbsp; :-}</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; -----Original Message-----</FONT>
<BR><FONT SIZE=3D2>&gt; From: powerh-l-admin@lists.sowder.com</FONT>
<BR><FONT SIZE=3D2>&gt; [<A =
HREF=3D"mailto:powerh-l-admin@lists.sowder.com">mailto:powerh-l-admin@li=
sts.sowder.com</A>]On Behalf Of Chris Sharman</FONT>
<BR><FONT SIZE=3D2>&gt; Sent: Wednesday, March 10, 2004 6:30 AM</FONT>
<BR><FONT SIZE=3D2>&gt; To: PowerH List</FONT>
<BR><FONT SIZE=3D2>&gt; Subject: odd PH/VMS logical name =
behaviour</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; I've got a mixed-case logical name:</FONT>
<BR><FONT SIZE=3D2>&gt; $ def &quot;Test&quot; =
&quot;MyTest&quot;</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; $ quiz</FONT>
<BR><FONT SIZE=3D2>&gt; Qz&gt; def mytest char*10 =3D =
getsystemval(&quot;Test&quot;)</FONT>
<BR><FONT SIZE=3D2>&gt; Qz&gt; rep mytest</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; Reports it as blank - presumably it doesn't =
find it.</FONT>
<BR><FONT SIZE=3D2>&gt; Defining it as uppercase works fine.</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; VMS logical names can be defined mixed-case =
(but are usually uppercase).</FONT>
<BR><FONT SIZE=3D2>&gt; They can be fetched either case-sensitive or =
case-blind.</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; Powerhouse would appear to be uppercasing the =
supplied string, then</FONT>
<BR><FONT SIZE=3D2>&gt; doing a case-sensitive search - not the optimum =
behaviour !</FONT>
<BR><FONT SIZE=3D2>&gt; Be nice if it used the string I gave it, and =
probably set the case-blind</FONT>
<BR><FONT SIZE=3D2>&gt; flag to avoid breaking most existing =
stuff.</FONT>
<BR><FONT SIZE=3D2>&gt; </FONT>
<BR><FONT SIZE=3D2>&gt; VMS 7.3, PH 7.10G3</FONT>
</P>
<BR>

<P><FONT =
SIZE=3D2>---------------------------------------------------------------=
--------</FONT>
</P>

<P><FONT SIZE=3D2>Any views expressed in this message are those of the =
sender and not</FONT>
<BR><FONT SIZE=3D2>necessarily those of CCA Group.&nbsp; The =
unauthorized use, disclosure,</FONT>
<BR><FONT SIZE=3D2>copying or alteration of this message is =
forbidden.&nbsp; The contents of</FONT>
<BR><FONT SIZE=3D2>this message may be confidential and/or privileged, =
copyright CCA Group</FONT>
<BR><FONT SIZE=3D2>and are intended solely for the use of the =
individual or entity to whom</FONT>
<BR><FONT SIZE=3D2>they are addressed.&nbsp; Whilst this message has =
been scanned, CCA Group</FONT>
<BR><FONT SIZE=3D2>cannot guarantee that it is virus free or compatible =
with your systems</FONT>
<BR><FONT SIZE=3D2>and accepts no responsibility for any loss or damage =
arising from its</FONT>
<BR><FONT SIZE=3D2>use. The recipient is advised to run their own =
anti-virus software. If</FONT>
<BR><FONT SIZE=3D2>you receive this message in error please =
contact</FONT>
<BR><FONT SIZE=3D2>postmaster@ccagroup.co.uk immediately, destroy any =
copies and delete it</FONT>
<BR><FONT SIZE=3D2>from your computer systems.</FONT>
<BR><FONT SIZE=3D2>=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =
=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D</FONT>
<BR><FONT SIZE=3D2>Mailing list: powerh-l@lists.sowder.com</FONT>
<BR><FONT SIZE=3D2>Subscribe: &quot;subscribe&quot; in message body to =
powerh-l-request@lists.sowder.com</FONT>
<BR><FONT SIZE=3D2>Unsubscribe: &quot;unsubscribe =
&lt;password&gt;&quot; in message body to =
powerh-l-request@lists.sowder.com</FONT>
<BR><FONT SIZE=3D2><A =
HREF=3D"http://lists.sowder.com/mailman/listinfo/powerh-l" =
TARGET=3D"_blank">http://lists.sowder.com/mailman/listinfo/powerh-l</A><=
/FONT>
<BR><FONT SIZE=3D2>This list is closed, thus to post to the list you =
must be a subscriber.</FONT>
</P>

<P><FONT SIZE=3D2>This message may contain privileged and/or =
confidential information.&nbsp; If you have received this e-mail in =
error or are not the intended recipient, you may not use, copy, =
disseminate or distribute it; do not open any attachments, delete it =
immediately from your system and notify the sender promptly by e-mail =
that you have done so.&nbsp; Thank you.</FONT></P>

</BODY>
</HTML>
------_=_NextPart_001_01C406DF.4F59FC2C--