RE: The target for a numeric operation is too small to hold the result

EVAL(R) was just an attempt,  as I couldn't find any other documentation
that would allow me to do it.

I have a   9,5   field   divided by  a (9,5 field/2000)   so the result
could have a large number of decimals,   I do not have to display this.
This is a cost per ton  so the field only needs to go out 2 decimal
places.  I keep getting the target error and am assuming it is because
my result field is also 9,5?   You are saying the eval shouldn't bomb
and should drop the decimals?

Once I created a work field of  25,21  I had no issues,  then moved this
field to my display field of 9,5.   So I believe dropping the decimals
greated than the 5th position must have been a problem.

Angela Wawrzaszek
IT Supervisor
Nucor Steel Auburn
(315) 258-4205
awawrzaszek@xxxxxxxxxxxxxxx

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Scott Klement
Sent: Friday, July 01, 2005 2:07 PM
To: RPG programming on the AS400 / iSeries
Subject: Re: The target for a numeric operation is too small to hold the
result


> How do I get around this error using the eval statement?  I DO NOT
want
> to make the field bigger,  I just want to truncate the extra decimals.

Huh? Do you mean that you want to truncate the extra digits?  Truncating

extra decimals is not necessary, it happens automatically.

For example, the following code won't receive any errors, in spite of
the 
fact that it drops a decimal place:

      D A               s              5P 1
      D B               s              5P 1
       /free
           a = 0.5;
           b = a / 2;

> It is a complex equation,  which I then push the result to a display
> field,  so to determine the max field size will be difficult,  plus I
> then have to move the 'temp' field to a display field.

If the max field size is a really big issue for you and you don't care 
about losing accuracy (which is very unsual for an RPG programmer, I
might 
add!) you might consider using floating point fields.  They allow for 
extremely huge numbers or extremely low numbers without a need to know 
ahead of time how many digits/decimals are needed.

The drawback is that they can lose accuracy. The result will be close,
but 
might not be exact.  Since you're not worried about digits being lost, I

assume that you don't care if your result is exact -- so this might be a

good solution.

> I just want to ignore the extra decimals!   I tried eval(r)    also
and
> still get the same errors

I don't understand what eval(r) has to do with your question.  Are you 
having trouble with an intermediate result?  If that's the case, you
could 
try using %DEC() to control the intermediate result size.