Re: gnubol: subsets

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

>>>>> "Bob" == RKRayhawk  <RKRayhawk@aol.com>
>>>>> wrote the following on Sat, 20 Nov 1999 20:26:04 EST

  Bob> In a message dated 11/18/99 8:51:23 AM EST, Michael McKernan
  Bob> mck@tivoli.mv.com writes:

  Bob> << There was a thread a while back on a similar ... subject,
  Bob> when Laura was in the midst of the Russian Armory experiment.
  Bob> 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


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

  Bob> I think that I am at long last getting to the END :-) of this
  Bob> thing and it looks a whole lot more reasonable to me now.  I
  Bob> think it is not political at all.  A way of looking at this is
  Bob> that the explicit scope terminators enable us to put anything
  Bob> within anything. That is in part what they are for. So a
  Bob> conditional statement becomes an 'imperative statement'
  Bob> because of the presence of the scope terminating element.
  Bob> That is really easy actually.

It's only not requiring them that's political, by which I simply mean
being influenced by other than technical considerations. 

  Bob> See, it is like this. You do not need to keep track of the
  Bob> fact that the inner arithmetic statement is within the
  Bob> conditional branch of some other arithmetic statement.  All
  Bob> you need to do is to define an explicitly terminated
  Bob> arithmetic statement as distinct from an arithmetic statement
  Bob> that is not explicitly terminated. That is much easier than
  Bob> doing anything like keeping track of the from within where
  Bob> issue.

Yup. That's what you have to do.  I don't think that's reasonable.
It means I get to (almost) duplicate every rule that admits of a
possible conditional phrase.   This is the tentative grammar I have for
ADD. 

add_stmt
	: ADD 
	  ( CORR identifier TO identifier { ROUNDED }
	  |  ( identifier | numeric )+
	     ( giving_phrase   	
	     ! TO ( identifier | numeric ) giving_phrase
	     | TO ( identifier { ROUNDED } )+
	     )
	  )
	  { size_err }
	  { END_ADD }
	;

giving_phrase
	: GIVING ( identifier { ROUNDED } )+
	;

size_err
	: {ON} SIZE {ERROR} statement
	  { ( NOT {ON} SIZE )? NOT {ON} SIZE ERROR (statement)+ }

	| NOT {ON} SIZE {ERROR} statement
	  { ( {ON} SIZE )? {ON} SIZE ERROR (statement)+ }
	;

You're asking me to add a rule similar to the following for every
verb that might include a conditional phrase.

imperative_add_stmt
	: ADD 
	  ( CORR identifier TO identifier { ROUNDED }
	  |  ( identifier | numeric )+
	     ( giving_phrase   	
	     ! TO ( identifier | numeric ) giving_phrase
	     | TO ( identifier { ROUNDED } )+
	     )
	  )
	  (  size_err END_ADD 
	  |  { END_ADD }
	  )
	;

and in size_error and other conditional sections, change <statement>
to a new rule <imperative>.  Maybe it could be done, but there's damn
little payoff for adding that much stuff, and I expect we'd have to
fight the tool in a number of places.

Perhaps my use of the terms "political" and "legalistic" sound a bit
too harsh, but it's hard not to see that they've left a loophole
solely for the purpose of not making existing code invalid.  There's
nothing especially wrong with that, but we don't have a floor and
ceiling standard, so I choose to compile all correct programs and a
few that are not.  I don't think that's especially wrong either.

BTW, you will note that size error is indifferent to the order of the
phrases, something that you seemed concerned about.  I do want to
compile all correct programs.

Mike






--
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.