Strange result in SORT Arithmatic Division

[Captain Nero]

Hi all,

I have a file in which the 17th and 18th positions have two numeric fields, I want to divide the first one by the second one and populate the 43rd position with a particular edit.

The card I have used in the OUTREC for this is -

OUTREC OVERLAY=(1,33,34:C'|RATIO = ',
43:(17,6,ZD,MUL,+10000000000),DIV,28,6,ZD,
EDIT=(IIIT.TTTTT),53:C'|')
SORT FIELDS=(1,2,CH,A)

However, I am getting some strange values in the 43rd column like sown.
AK - 000001|F - 000000|M - 000001|RATIO = 0.00000|
AL - 000029|F - 000014|M - 000015|RATIO = 3571.42857|
AR - 000013|F - 000005|M - 000008|RATIO = 1700.68027|
AZ - 000032|F - 000012|M - 000020|RATIO = 1581.02766|
CA - 000242|F - 000100|M - 000142|RATIO = 975.60975|
CO - 000028|F - 000012|M - 000016|RATIO = 1403.50877|
CT - 000070|F - 000030|M - 000040|RATIO = 833.33333|

Can somebody please explain me why is this coming and what can be the remedy. I have multiplied the first field by 10000000000 as this was a requirement.
_________________
Regards,
Captain N.

[Frank Yaeger]

I don't know how you got

3571.42857

from 14 and 15 with that statement. When I do it with DFSORT, I get:

3333.33333

which is what I'd expect. Here's why:

DFSORT does integer arithmetic (it throws away the decimal places), so:

(14 * 10000000000)/15 = 9333333333

There are 10 digits in that result but your edit mask of IIIT.TTTTT only has 9 digits, so the digits are mapped right to left and the first digit (the 9 gets dropped). The result is:

3333.33333

If you had 10 digits in your edit mask as IIIIT.TTTTT, the leading digit wouldn't be dropped and you'd get:

93333.3333

But I don't know where you got

3571.42857

You'd have to prove to me that DFSORT produced that result with the input you showed.

By the way, since you're using OVERLAY, you don't need 1,33 - OVERLAY will leave that data as is.
_________________
Frank Yaeger - DFSORT Development Team (IBM) - yaeger@us.ibm.com
Specialties: JOINKEYS, FINDREP, WHEN=GROUP, ICETOOL, Symbols, Migration
=> DFSORT/MVS is on the Web at http://www.ibm.com/storage/dfsort/

[Captain Nero]

Hi Frank,

I have put the result directly from the mainframe. However, i will get you the DFSMSG on Monday. I am really quite perplexed to see the results.

Hope something will turn out.
_________________
Regards,
Captain N.

[Frank Yaeger]

Well, at the very least you should make the edit mask larger so you can see the full result rather than a truncated result.

It would be helpful to "display" the relevant input fields and the actual output values you received for those input fields with and without truncation, all in the same line, as well as the DFSORT messages.
_________________
Frank Yaeger - DFSORT Development Team (IBM) - yaeger@us.ibm.com
Specialties: JOINKEYS, FINDREP, WHEN=GROUP, ICETOOL, Symbols, Migration
=> DFSORT/MVS is on the Web at http://www.ibm.com/storage/dfsort/

[Captain Nero]

Yes frank I will make the changes on the mask and the Overlay of the first 33 bytes as you have suggested, however, the input file and the output are right in front of you... My input was without the last column and I have added the last column(Ratio) in the output.

Thanks.
_________________
Regards,
Captain N.

[Frank Yaeger]

Sanity check. Let's simplify things.

When I run this with DFSORT:

[Captain Nero]

Frank I also got output as expected the problem was with an INREC statement which I have put in different control.

Now it is working fine.
_________________
Regards,
Captain N.

[Captain Nero]

And Thank you so much for the help.
_________________
Regards,
Captain N.

[Frank Yaeger]

You're welcome. Glad to hear it was a "user error". Thanks for letting us know. .
_________________
Frank Yaeger - DFSORT Development Team (IBM) - yaeger@us.ibm.com
Specialties: JOINKEYS, FINDREP, WHEN=GROUP, ICETOOL, Symbols, Migration
=> DFSORT/MVS is on the Web at http://www.ibm.com/storage/dfsort/