IBM Mainframe Forum

Posted: **Mon Dec 17, 2012 6:55 pm**

Hi All,

I need help with below Sort:
Input data will be of this format:

"1234"||"567"||"gtyhjgdgh"||" " ||" " ||"y"||
"4"||"5"||"hj8bjh"||" "||" " ||"y"||
"4"||"6"||"hj8bjh"||" "||" " ||"y"||
"1"||"7"||"98bjh"||" "||" " ||"y"||
"9"||"6"||"gtyhh"||" " ||"Y " ||"y"||
"4"||"4"||"hj8bjh"||" "||" " ||"y"||

and o/p should be

Select all

1234
4
1
9

i.e. it should eliminate the " and || and should also stop copying when it hits | in the first line and move to next line.

With FINDREP option I am able to remove " and || but how to stop copying once it has encountered | and move copy to next line.

//STEP0001 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=FILE1,DISP=SHR
//SORTOUT DD DSN=FILE2,DISP=SHR
//SYSIN DD *
OPTION COPY
INREC FINDREP=(INOUT=(C'"',C'',C'|',C''))

is giving me

Select all

1234567GTYHJGDGH Y
45HJ8BJH Y
46HJ8BJH Y
1798BJH Y
96GTYHH Y Y
44HJ8BJH Y

which is working as FINDREP is coded.

Regards..

Posted: **Mon Dec 17, 2012 9:09 pm**

Why nor use PARSE ? If it supports your level

Posted: **Mon Dec 17, 2012 9:12 pm**

Hi All,

A small update:

I used PARSE to edit the data required, ( have removed quotes from i/p file using FINDREP in STEP001)...this is
working fine...

Select all

//STEP0002 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=FILE1,DISP=SHR
//SORTOUT DD DSN=FILE2,DISP=SHR
//SYSIN DD *
INREC PARSE=(%01=(ENDBEFR=C'|',FIXLEN=16)),
BUILD=(1:%01)
SORT FIELDS=(1,16,FI,A)
SUM FIELDS=NONE
//*

Where File1 looks like

Select all

----+----1----+----2----+----3----+----4-
1234||567||GTYHJGDGH|| || ||Y||
4||5||HJ8BJH|| || ||Y||
4||6||HJ8BJH|| || ||Y||
1||7||98BJH|| || ||Y||
9||6||GTYHH|| ||Y ||Y||
4||4||HJ8BJH|| || ||Y||

and File 2 as

Select all

----+----1---
1
1234
4
9

However, the SORT FIELDS=(1,16,FI,A) is not working as I expected..
The result I was expecting was
1
4
9
1234

and am getting is
1
1234
4
9

Kindly correct me where I am going wrong.

Posted: **Mon Dec 17, 2012 11:25 pm**

Hello,

Suggest you use UFF.

Posted: **Tue Dec 18, 2012 4:19 am**

use the following DFSORT JCL which will give you the desired results

Select all

//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD *
"1234"||"567"||"GTYHJGDGH"||" " ||" " ||"Y"||
"4"||"5"||"HJ8BJH"||" "||" " ||"Y"||
"4"||"6"||"HJ8BJH"||" "||" " ||"Y"||
"1"||"7"||"98BJH"||" "||" " ||"Y"||
"9"||"6"||"GTYHH"||" " ||"Y " ||"Y"||
"4"||"4"||"HJ8BJH"||" "||" " ||"Y"||
//SORTOUT DD SYSOUT=*
//SYSIN DD *
INREC PARSE=(%01=(STARTAFT=C'"',ENDBEFR=C'"',FIXLEN=16)),BUILD=(%01)
SORT FIELDS=(1,16,UFF,A)
SUM FIELDS=NONE
//*

Posted: **Tue Dec 18, 2012 12:46 pm**

Thanks Kolusu .. It worked ! I got my mistake :-/
Cheers

IBM Mainframe Forum

Copying Distinct Records

Copying Distinct Records

Re: Copying Distinct Records

Re: Copying Distinct Records

Re: Copying Distinct Records

Re: Copying Distinct Records

Re: Copying Distinct Records