IBM Mainframe Forum

by **Nary** » Fri Oct 09, 2009 10:00 am

Hi,

I have a dataset which includes all the original records and duplicate records.
How do I separate original records and duplicate records and place them in two different datasets?

Thank you.
Nary.

by **Balamurugan3** » Fri Oct 09, 2009 1:27 pm

Hi

think so this would be useful to you... some time back i was too searching for the same solution..... I got that from another forum...This is using utility ICETOOL ...

Select all

//STEP1 EXEC PGM=ICETOOL,REGION=1024K
//TOOLMSG DD SYSOUT=*
//DFSMSG DD SYSOUT=*
//IN DD DSN=XXX.YYY.INPUT,DISP=SHR
//SORTXSUM DD DSN=XXX.YYY.DUP,DISP=SHR
//OUT DD DSN=XXX.YYY.OUTPUT,DISP=SHR
//TOOLIN DD *
SELECT FROM(IN) TO(OUT) ON(1,3,CH) FIRST DISCARD(SORTXSUM)
/*

Here i have mentioned it with use of 1 key.. if so if your file has more than 1 key...

use it like

ON(1,3,CH) ON(4,2,CH)...FIRST DISCARD(SORTXSUM)

IBM Mainframe Forum

Splitting records using SORT

Splitting records using SORT

Re: Splitting records using SORT