RE: gnubol: Compiler Limits

["William M. Klein" <wmklein@ix.netcom.com>]

> -----Original Message-----

>
> Are there other aspects of limits that we should think about before coding
> syntax / semantics so that we can see ahead of time where
> parametric devices
> might be considered early?
>

Some areas that various compilers set limits on.  (The ANSI/ISO Standard is
*quite* clear that it is up to the implementor what limits they place on
"ellipsis" items - so any limit is "legal" whether or not it is
user-friendly)

1) Size of Alphanumeric item (eg.  PIC X(999999999...))

2) Size of tables (with or without OCCURS DEPENDING ON) - some
implementations allow larger "fixed length" tables than they doing variable
length tables.

3) Number of REPLACING phrases in COPY/REPLACING (also REPLACE verb)

4) Number of REPLACING phrases in INSPECT

5) Number of items in "INDEXED BY" (in a single OCCURS clause)

6) Already mentioned in one thread - total number of lines in a compilation
unit (which can include either nested or series of programs)

7) Number of items in GO TO DEPENDING ON (YUCK !!! - but it is in the '85
Standard)

  ****

Someplace I have a document that was created by the MIA consortium (a
Japanese COBOL definition) which tells which items they places limits on.
Similarly, the old IBM SAA definition provided some limits.  I will look for
those documents (hardcopy only) and see what I find.

Bill Klein
  wmklein <at> ix.netcom.com


--
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.