Re: gnubol: Free-form problem

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

>>>>> "Tim" == Tim Josling <tej@melbpc.org.au>
>>>>> wrote the following on Mon, 15 Nov 1999 05:43:47 +1000

  Tim> Basically this can't be cleanly handled. In the preprocessor I
  Tim> go into a special lex mode, where I scan ahead for:

  Tim> a) Anything in area a ("warning - comment entry assumed ended
  Tim> by next paragraph") b) Period space.

  Tim> Don't forget that continuations are not allowed in comment
  Tim> entries, and lots of other trash. That's why I did the PP to
  Tim> handle all this ugly stuff.

  Tim> The preprocessor basically throws away the comment entries,
  Tim> except for date-compiled where it retains only the paragraph
  Tim> header.

Gee, I'm glad you're taking care of this, Tim ;-)

Even so, the language in the original standard makes these paragraphs
completely dependent on syntactic margin A.  If we're going to have
free form COBOL, some accommodation must have been made.  I'm still
curious about how it's handled.  If I write something like this, how
much trouble am I in?

     IDENTIFICATION DIVISION.
     PROGRAM-ID. FOO.
     AUTHOR. ENVIRONMENT DIVISION.
     INSTALLATION. DATA DIVISION.
     DATE-WRITTEN. PROCEDURE DIVISION.

     ENVIRONMENT DIVISION.

Even "PERIOD SPACE" is not sacrosanct by the language in my book.
Since the comment-entry is permitted to occupy multiple line, if we
go strictly by the book and ignore margin A, the next thing we'll see
is eof.  I doubt that's what is intended.



--
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.