RE: Rounding up

For a number with 2 decimal places the half adjust adds 0.005 to the result
and then takes the result to the second decimal place, so 1.234 would round
to 1.23 and 1.237 would round to 1.24.

To have your result round up to the next whole number regardless you need to
add 0.99 to your result and take the integer, so in your examples you would
have:

(2 * 1.01) + 0.99 = 2.02 + 0.99 = 3.01 = 3
(2 * 1.50) + 0.99 = 3.00 + 0.99 = 3.99 = 3
(2 * 1.51) + 0.99 = 3.01 + 0.99 = 4.00 = 4

All the best

Jonathan



-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Lance Gillespie
Sent: 27 February 2009 22:39
To: rpg400-l@xxxxxxxxxxxx
Subject: Rounding up

I have a situation where I need my results rounded up if they are not exact.

As in:
2 * 1.01 = 3
2 * 1.50 = 3
2 * 1.51 = 4

The source and target numbers are 7,0 and the factor is a constant.

I thought:

target = %dech(source * 1.01) : 7 : 0)

or maybe

target = %dech(source * 1.01) : 9 : 2)

would do it, but no joy.

I suppose I could do the calcs twice with %dec, once to a 7,0 result and
once to a 9,2 result
(assuming %dec truncates) and then add one to the 7,0 result if the two are
not equal, but
I would rather not.

Is there some way to force these calcs to round up if not exact?

Thanks,
Lance