IBM Mainframe Forum

Posted: **Sat Sep 20, 2014 3:02 pm**

Hi all,

I would like to know how to split records using JCL.
Scenario is below,

I have a dataset which has students records of each exam like,

Select all

JAY ANNUALEXAM 7.9 FIRSTCLASS
RAM QUARTEREXAM 7.0 FIRSTCLASS
RAM HALFYRLYEXAM 7.6 FIRSTCLASS
RAM ANNUALEXAM 8.3 FIRSTCLASS
SAM QUARTEREXAM 6.5 FIRSTCLASS
SAM ANNUALEXAM 7.3 FIRSTCLASS
RAJ QUARTEREXAM 6.7 FIRSTCLASS
RAJ HALFYRLYEXAM 5.6 SECONDCLASS
DAV ANNUALEXAM 6.4 FIRSTCLASS

the above nine records are in one single dataset. I need to split it into 2 records by using col(1,3) as key in runtime.
The key value should not be hard coded in the sort card.

The output files should like below,

File 1:

Select all

JAY ANNUALEXAM 7.9 FIRSTCLASS
RAM QUARTEREXAM 7.0 FIRSTCLASS
RAM HALFYRLYEXAM 7.6 FIRSTCLASS
RAM ANNUALEXAM 8.3 FIRSTCLASS

File 2:

Select all

SAM QUARTEREXAM 6.5 FIRSTCLASS
SAM ANNUALEXAM 7.3 FIRSTCLASS
RAJ QUARTEREXAM 6.7 FIRSTCLASS
RAJ HALFYRLYEXAM 5.6 SECONDCLASS
DAV ANNUALEXAM 6.4 FIRSTCLASS

Regards,
Dhamodharan K

Code'd

Posted: **Sat Sep 20, 2014 4:06 pm**

Have a look at WHEN=GROUP with KEYBEGIN and PUSH an ID (a sequence number which only changes when the group definer changes) then use OUTFIL with INCLUDE= for the ID field being less-than-or-equal-to two.

Posted: **Sat Sep 20, 2014 4:20 pm**

Hi Billy,

Could you please elaborate more.

Regards,
Dhamodharan K

IBM Mainframe Forum

Splitting files using sort in JCL

Splitting files using sort in JCL

Re: Splitting files using sort in JCL

Re: Splitting files using sort in JCL