Re: gnubol: subsets

[Michael McKernan <mck@tivoli.mv.com>]

>>>>> "Tim" == Tim E Josling <TIMJOSLING@prodigy.net>
>>>>> wrote the following on Thu, 18 Nov 1999 14:48:15 +1000

  Tim> The standard specifically requires us to flag extensions upon
  Tim> request. I think we should try and do this. Someone would have
  Tim> to write code to find these and warn the user. As an option
  Tim> you can turn on/off. I will research further how you can pick
  Tim> it up.

  Tim> Tim Josling

  Tim> Michael McKernan wrote:

  >> I didn't even notice that grammar problem; I just tried to get
  >> the original grammar to go through an LL/1 parser... Right now,
  >> I think the best thing to do is accept it.

Anything is possible, of course, but I hesitate to add otherwise
superfluous scaffolding to our processor to account for what I
consider political decisions of the committee.  As you are quite
aware, the committee is extremely reluctant to remove a language
element until n generations of the standard after they have provided
an adequate replacement for it.  The NEXT SENTENCE phrase will
undoubtedly be deprecated eventually, and perhaps removed when doing
so does not create a storm of protests.  In the meantime, the
committee discourages its use by legalistically prohibiting its use
in the same context as the new feature.  

There was a thread a while back on a similar similar subject, when
Laura was in the midst of the Russian Armory experiment.  As I
remember it went something like this.

	 ADD a TO b
	 ON SIZE
	    <many statements>
	    SUBTRACT c from d
	    ON SIZE
	       <many more statements>
	 END ADD

Now, the SUBTRACT is illegal because it is not an imperative
statement, not having an END SUBTRACT.  The imperative statement
requirement is the same kind of legalistic prohibition.  Accepting
the fragment as written causes no difficulty in parsing, and its
meaning is clear, but it appears to require an extraordinary effort
to diagnose because you have to be aware that the SUBTRACT is in the
scope of a construct that does not allow conditional statements when
you arrive at the END ADD.

I am not purist enough to want to complicate our task substantially
to catch his class of error, but if the group thinks it's important,
I'll spend some time on it.  Is it significant that, when
certification of COBOL compilers was required, the audit tests never
probed this area?  Others on this list may be able to speak more
authoritatively on committee matters, but it does sometimes appear
that they are willing to wink at certain infractions rather than
discuss whether their requirements make unreasonable demands on the
implementor.





--
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.