[comp.os.vms] Did I Find a Bug?!?

ACSCCRA@ubvms.BITNET ("Curtis R. Anderson") (06/20/87)

On VAX/VMS 4.5, I discovered this (when I was really paying attention):

$ SHOW LOGICAL FOO
%SHOW-S-NOTRAN, no translation for logical name FOO

I hardly think that was "sucessful;" more like it should be "%SHOW-W-NOTRAN"...

What do you think?  Should I get our university to SPR it or does somebody in
net/newsgroupland know if SHOW.EXE has been fixed, and we never got it?

I had to quickly check another machine in the cluster to see that it was there,
too...

+------------------------------------------+-----------------------------------+
|  Curtis R. Anderson                      |                                   |
|  UCS Student Consultant                  | Syntax:                           |
|                                          |                                   |
|  ACSCCRA@UBVMS.BITNET                    |   A levy placed on wrongdoing.    |
|  ACSCCRA@UBVMSC.CC.BUFFALO.EDU           |                                   |
|  SnailNet:                               |                                   |
|  State University of New York at Buffalo |                                   |
|  220 Computing Center                    |                                   |
|  Buffalo, NY 14260 USA                   |                                   |
+------------------------------------------+-----------------------------------+

u3369429@murdu.OZ (Michael Bednarek) (06/22/87)

In article <8706210336.AA18429@ucbvax.Berkeley.EDU> ACSCCRA@ubvms.BITNET ("Curtis R. Anderson") writes:
>$ SHOW LOGICAL FOO
>%SHOW-S-NOTRAN, no translation for logical name FOO
>
>I hardly think that was "sucessful;" more like it should be "%SHOW-W-NOTRAN"...
>
>What do you think?
 
Well, at least you got a $STATUS returned (%X10788019 I believe).

However, consider this:

You run two programs in a batch procedure. The first one reads some lines
from SYS$INPUT. It may discover that there is an error in your input,
inform you duly what it is, and exit. Leaving some more input records
in the procedure unread. Then DCL will come back and issue:
%DCL-W-SKPDAT, image data (records not beginning with "$") ignored

In anticipation of that, you have a check in the procedure:
$ On Warning then Exit
which DOESN'T exit. Why? I don't know.
It seems that the error message DCL-W-SKPDAT doesn't affect $STATUS, to wit:

$ directory foo
%DIRECT-W-NOFILES, no files found
foo bar foo
%DCL-W-SKPDAT, image data (records not beginning with "$") ignored
$ Show Symbol $STATUS
  $STATUS == "%X10018290"

(The $STATUS shown is a result from the DIRECTORY command.)

Can anyone enlighten me?

Michael Bednarek (u3369429@murdu.oz.au)

P.S.: The work-around is that I educated my users to write programs that
      exit with a meaningful status, like SS$_NORMAL or SS$_ABORT.
      But I can't do that with existing programs.