Dumb "figurative constant" rules
- From: "William M. Klein" <wmklein@xxxxxxxxxxxxxxxxx>
- Date: Wed, 25 May 2005 21:26:36 GMT
Just as an (amusing???) follow-up to the current discussion on HIGH-VALUES and
LOW-VALUES, I thought I would mention the following:
The '85 Standard (and I assume all earlier ones) allows one to MOVE
- QUOTE
- HIGH-VALUES
- LOW-VALUES
- all "literal" *> with alphanumeric characters in the literal
to a NUMERIC (any usage - I think) receiving field - in SOME very rare cases
LOW-VALUES *might* be a "DISPLAY" zero or HIGH-VALUES a display nine, but I
suspect this is INCREDIBLY rare. When QUOTE is a numeric integer, is even more
problematic.
The '02 Standard made this "archaic" (not obsolete) and the draft of the '08
Standard gives some more examples of when this would work - but not as expected,
e.g.
05 Num1 Pic 99V99
...
Move All "9" to Num1
Evaluate Num1
when 99.99
Display "You might think so - but NO"
when 99.00
Display "This is standard - but is it expected?"
End-Evaluate
P.S. The '02 Standard introduced "Function Highest-Algebraic" (and
Lowest-Algebraic) to give the "expected" results for the above code.
--
Bill Klein
wmklein <at> ix.netcom.com
.
- Prev by Date: Re: Occurs Depending Memory Use
- Next by Date: Re: In Cobol, large amount of spaces between records due to variable definition
- Previous by thread: Linking Problem
- Next by thread: Where are spaces required in COBOL source code?
- Index(es):
Relevant Pages
|
