[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: gnubol: How do we parse this language, anyway?



In a message dated 12/6/99 10:36:59 AM EST, david.ondrejko@safelite.com 
writes:

<< 
 Unfortunately, I'm working with COBOL '74 under DOS/VSE, which doesn't
 include the NOT SIZE ERROR clause.  (It also worries about the lack of
 an ENVIRONMENT DIVISION.)  If you have any questions about backward
 compatibility, I might be able to provide some input then...
 
 -- 
 David R. Ondrejko - >>


I have a question about simple nesting. Did the 'old' compiler which you are 
still currently using allow arithmetic statements on the end of conditionals 
at all.  As in


   ADD a TO b
       ON SIZE ERROR
            ADD 1 TO  error-count
   .

Did the old compiler allow nesting of conditionals to the second level

   ADD a TO b
       ON SIZE ERROR
            ADD 1 TO  error-count
               ON SIZE ERROR
                  ADD 1 TO emergency-counter 
   .

Do we get warnings, silence, rejects; What?

Without wanting to put demands on your time in terms of research, from your 
experience alone is there any difference in rules if the conditionals are 
nested inside of other conditionals? As in 

   READ ....
       AT END
          ADD 1 to crazy-eof-counter
.

Where crazy-eof-conter might be the means to control PERFORMS, who knows. 

Or more dramatically

   READ ....
       AT END
          ADD 1 to crazy-eof-counter
             ON SIZE ERROR
                DISPLAY 'we must have been here a large number of times'
.


These second two are after any differences in the liberalism of the compiler 
if it allows some
conditionals in conditionals (if not then most of it errors out).

There isn't one of these I would recommend. I am actually just making a 
little more hunt for any mainframe model that may have lead PC or midi 
products to try to accept inner conditionals as positive extension.  
Increasingly, I think those products actually just fell victim to an easy 
solution. ((which I am guessing becomes truely hard to support under 
transposition leniencies in the future)).

Your time is valued. I think just a few combinations will suffice.

Thanks
Bob Rayhawk





             


  




--
This message was sent through the gnu-cobol mailing list.  To remove yourself
from this mailing list, send a message to majordomo@lusars.net with the
words "unsubscribe gnu-cobol" in the message body.  For more information on
the GNU COBOL project, send mail to gnu-cobol-owner@lusars.net.