OUTFIL: Splitting
OUTFIL: Splitting
Summary
- OUTFIL can be used to split the dataset with a specific number of records. SPLIT key word used to split the dataset.
- If SPLIT is used and there are two OUTFIL datasets, the SORT will distribute first record to OUTFIL dataset1 and second record to OUTFIL dataset2 and third to OUTFIL dataset1 and so on until the end of the records.
- If SPLIT=n used and two OUTFIL files, then n records will be distributed from first OUTFIL dataset, the next n records to OUTFIL dataset2, the next n records to OUTFIL file1 and so on until the end of the records.
- If the number of files changed, then the process will adjust accordingly. SPLIT splits the records one ata time among the data sets specified by FNAMES.
Syntax -
OPTION COPY
OUTFIL FNAMES=(DDNAME1, DDNAME2,….,DDNAMEn), SPLIT=n
DDNAMEn | DDNAME is eight character’s name that representing the actual dataset in JCL. |
SPLIT=n | Specifies the number of records that the file needs to split with. In SPLIT=n, n is optional. If n is not specified, the file split up can be done record by record. If n is specified, the file split up can be done n records by n records. |
Example -
Scenario - Create three output files by splitting 4 records each.
Input File - MTHUSER.SORT.INPUT01 - FB file of 80 length
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
********************************* Top of Data **********************************
00001 student1 dept1 560
00003 student3 dept2 520
00004 student4 dept1 540
00005 student5 dept2 500
00002 student2 dept3 510
00006 student6 dept3 550
00008 student8 dept1 530
00007 student7 dept3 510
00009 student9 dept2 520
00010 student10 dept2 505
******************************** Bottom of Data ********************************
JCL -
----+----1----+----2----+----3----+----4----+----5----+----6----+----7--
***************************** Top of Data ******************************
//Job-card
//*
//**********************************************************************
//*
//STEP01 EXEC PGM=SORT
//SORTIN DD DSN=MTHUSER.SORT.INPUT01,DISP=SHR
//OUTPUT1 DD DSN=MTHUSER.SORT.OUTPT01,
// SPACE=(CYL,(1,1),RLSE),DCB=*.SORTIN,
// DISP=(NEW,CATLG,DELETE)
//OUTPUT2 DD DSN=MTHUSER.SORT.OUTPT02,
// SPACE=(CYL,(1,1),RLSE),DCB=*.SORTIN,
// DISP=(NEW,CATLG,DELETE)
//OUTPUT3 DD DSN=MTHUSER.SORT.OUTPT03,
// SPACE=(CYL,(1,1),RLSE),DCB=*.SORTIN,
// DISP=(NEW,CATLG,DELETE)
//SYSOUT DD SYSOUT=*
//SYSIN DD *
SORT FIELDS=(1,5,ZD,A)
OUTFIL FNAMES=(OUTPUT1,OUTPUT2,OUTPUT3),SPLITBY=4
/*
**************************** Bottom of Data ****************************
Output File1 - MTHUSER.SORT.OUTPT01 - FB file of 80 length
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
********************************* Top of Data **********************************
00001 student1 dept1 560
00002 student2 dept3 510
00003 student3 dept2 520
00004 student4 dept1 540
******************************** Bottom of Data ********************************
Output File2 - MTHUSER.SORT.OUTPT02 - FB file of 80 length
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
********************************* Top of Data **********************************
00005 student5 dept2 500
00006 student6 dept3 550
00007 student7 dept3 510
00008 student8 dept1 530
******************************** Bottom of Data ********************************
Output File3 - MTHUSER.SORT.OUTPT03 - FB file of 80 length
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
********************************* Top of Data **********************************
00009 student9 dept2 520
00010 student10 dept2 505
******************************** Bottom of Data ********************************
Explaining Example -
- SORT FIELDS=(1,5,ZD,A) - sort all outputs in ascending order based on the ZD value from 1 to 5 positions.
- OUTFIL FNAMES=(OUTPUT1,OUTPUT2,OUTPUT3),SPLITBY=4 - Splits the records by 4. First 4 records to output1 file, next 4 records to Output2 file and next 4 to output3 file.
- If any more records than 12, then the above process will occur again and again until it reaches to file end.