IBM Mainframe Forum

[naveenkumar sudha]

Hi Team,

I have an FB file as below, consisting of 5 fields - A,B,C,D(YEARMM),E(Amount), delimited by pipe symbol
I need to sum-up amounts and count no of records, for all matching A,B,C,D(YEARMM) and write to a new file having fields- A,B,C,D(YEARMM), accumulated/summarized AMOUNT and no of records having same A,B,C,D.

I/P:
----+----1----+----2----+----3----+----4----+----5----+----
C1AW|1013300 |16L00000 |202201|0001
C1AW|1013300 |16L00000 |202202|0001
C1AW|1013300 |16L00000 |202202|0001
C1AW|1020000 |16L21000 |202112|0001
C1AW|1020000 |16000000 |202103|0001
C1AW|1020000 |16000000 |202103|0002
C1AW|1020000 |16000000 |202106|0001
O/P Expected:
-----------------
C1AW|1013300 |16L00000 |202201|0001|1
C1AW|1013300 |16L00000 |202202|0002|2
C1AW|1020000 |16L21000 |202112|0001|1
C1AW|1020000 |16000000 |202103|0003|2
C1AW|1020000 |16000000 |202106|0001| 1

I can sum-up amounts based on column:A,B,C,D using SUM FIELDS=NONE and use M10 function to count records, having individual steps. But again, i need to merge these individual steps which is inefficient.
Hence i would need your help, can this be done in single step - both summing up and writing count of records based on matched key.

Kindly help.

Thanks a lot,
Naveen S

[sergeyken]

0. Please, learn how to use the Code button to format your code properly

1. Please understand that you cannot do your job using JCL. You need to use any SORT utility

2. Please read the Dividing a Report into Sections part of SYNCSORT manual (Page 3.42). Then, try to use this new information in your task.

3. Finally, present you results at this forum, if you still have extra questions.

[sergeyken]

2.1. In case you are using DFSORT utility, please read SECTIONS part of the manual (Page 339).

[naveenkumar sudha]

Sorry, I have logged in to the forum after a long time. I have just realized I should have posted this in DFSORT group. Thank you.