IBM Mainframe Forum

by **vegafacundodaniel** » Tue Jul 20, 2010 4:41 pm

I have an input VARIABLE dataset. Something like this :
-TEST1
-123456
-ABCDEF
-BFDSS

My field starts at the 1st position of the file and it takes 6 positions. X(6)

I must sort all records with :
- For the records with 6 characters in the field, I have to take the last 5 characters. For ex: For ABCDEF, I need to take BCDEF.
- For the records with 5 characters in the field, I have to take the 5 characters. For ex: For TEST1, I need to take TEST1

The sorted output file must be:
-BCDEF
-BFDSS
-TEST1
-23456

It is important to consider that the input file is variable.

by **skolusu** » Tue Jul 20, 2010 9:24 pm

vegafacundodaniel wrote:It is important to consider that the input file is variable.

vegafacundodaniel,

Did you mean to say that the input file is DCB is VB which has the RDW in the first 4 bytes? or the file is FB file but the DATA in the file is variable?

Since you did not provide the LRECL and RECFM , for now I am going to assume it is FB recfm and LRECL of 80.

The following DFSORT JCL will give you the desired results

Select all

//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD *
-TEST1
-123456
-ABCDEF
-BFDSS
//SORTOUT DD SYSOUT=*
//SYSIN DD *
INREC OVERLAY=(81:2,6,JFY=(SHIFT=RIGHT))
SORT FIELDS=(82,5,CH,A)
OUTREC BUILD=(1,1,82,5,X,8,73)
//*

The output of this is

Select all

-BCDEF
-BFDSS
-TEST1
-23456

by **vegafacundodaniel** » Wed Jul 21, 2010 4:36 pm

Thanks Skolusu for your help,

My input file is VB of 1000 characters (which has the RDW in the first 4 bytes) and my output file is FB 996.

Thanks in advance !!!

by **skolusu** » Wed Jul 21, 2010 11:20 pm

vegafacundodaniel wrote:Thanks Skolusu for your help,

My input file is VB of 1000 characters (which has the RDW in the first 4 bytes) and my output file is FB 996.

Thanks in advance !!!

*Sigh* You show an example of 6 characters out of which you removed just 1 character . But now you show that need 4 bytes to be taken off from the input. What is the position and format of the input field that you want to sort and what are the complete rules?

Here is a sample job which assumes that your sorting start starts at pos 6 for 6 bytes and you want to remove the first byte of it. The output file is VB with LRECL of 999

Select all

//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=Your input VB 1000 byte file,DISP=SHR
//SORTOUT DD SYSOUT=*
//SYSIN DD *
INREC BUILD=(1,4,6,6,JFY=(SHIFT=RIGHT),5)
SORT FIELDS=(6,5,CH,A),EQUALS
OUTREC BUILD=(1,4,11,1,6,5,18)
//*

by **NicC** » Thu Jul 22, 2010 1:08 am

errr...Kolusu - he wants a RECFM=FB LRECL=996 output file

by **skolusu** » Thu Jul 22, 2010 2:01 am

NicC wrote:errr...Kolusu - he wants a RECFM=FB LRECL=996 output file

oops . just noticed that. Thanks, In that case he needs to use the following control cards.

Select all

//SYSIN DD *
INREC BUILD=(1,4,6,6,JFY=(SHIFT=RIGHT),5)
SORT FIELDS=(6,5,CH,A)
OUTFIL VTOF,BUILD=(11,996)
//*

IBM Mainframe Forum

Sort Depending on Field Length

Sort Depending on Field Length

Re: Sort Depending on Field Length

Re: Sort Depending on Field Length

Re: Sort Depending on Field Length

Re: Sort Depending on Field Length

Re: Sort Depending on Field Length