JCL to count the number of duplicate records in a dataset

Say you have an input file with transaction details and want to count the number of records with same transaction ID.

 

Input File

TRN1

TRN1

TRN1

TRN2

TRN2

TRN3

 

Output File

TRN1  3

TRN2  2

TRN3  1 

 

JCL

//STEP10 EXEC PGM=ICETOOL
//TOOLMSG DD SYSOUT=*
//DFSMSG DD SYSOUT=*
//OUTFILE DD DSN=TRNCNT.PS,DISP=SHR

//INFILE DD DSN=TNFIL.PS,DISP=SHR

//TOOLIN DD *

OCCUR FROM(INFILE) LIST(OUTFILE) -
NOHEADER NOCC-
ON(1,4,CH) -
ON(VALCNT,N06)
/*

 

 

The duplicate values will be looked only in the first 4 bytes of each record and count will be displayed in the 6th position. 

 

If you want compare records with 80 bytes change 4 as 80 and 6 to the position where you want to display your count