Re: gnubol: WHEN and other conditional phrases

[Mike Mckernan <mckernan@const.zko.dec.com>]

>>>>> "Bob" == RKRayhawk@aol.com
>>>>> wrote the following on Wed, 15 Dec 1999 10:11:09 -0500 (EST)

  Bob> In a message dated 12/15/99 8:44:41 AM EST, mck@tivoli.mv.com
  Bob> writes:

  Bob> Lots of snipping ...  << Essentially, we're parsing as if the
  Bob> terminators are not required, making diagnosis an incidental
  Bob> requirement.
  >>>

  Bob> I think that is an excellent view.  IMHO it leads to tearing
  Bob> the grammar rules into fragments, and managing considerable
  Bob> portions of scope issues in action code.

Pro forma disagreement to be noted, of course.

  Bob> (putting asside code base issues for _sequential_ WRITES with
  Bob> INVALID KEY clauses).  If we do not manifest type then

READ RANDOM-FILE
     INVALID KEY
        WRITE SEQUENTIAL-FILE
     NOT INVALID KEY

  Bob> can not be parsed as discussed by means of grammar rules,
  Bob> because you are not allowed to reduce that I/O verb in front
  Bob> of the conditional clause (NOT INVALID KEY) as a simple
  Bob> imperative, any more than arithmetics can be reduced as simple
  Bob> imperatives infront of a conditional clause (NOT ON SIZE
  Bob> ERROR).

You're right, Bob.  This assumes facts not in evidence.  It is only
possible to parse this as I described with some context awareness.
Context does not help with the arithmetics, though.

Best regards,

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.