1) Compare two files & write matching records to output file.
Option1:
This program will degrade the performance but for understanding the logic this option has given. Option2 is efficient. It is explained below option1 program.
IDENTIFICATION DIVISION.
JCL to Execute above cobol program -
//STEP010 EXEC IGYWCLG,MEM=FILCMPR
//SYSPRINT DD SYSOUT=*
//COBOL.SYSIN DD DSN=XX7384A.JCLLIB.SELFTEST.Y2013.COB(&MEM),
// DISP=SHR
//GO.DD1 DD *
21400 SANDIP 33333
21500 MANDIP 44444
21600 RANDIP 99999
21700 BALDIP 66666
//GO.DD2 DD *
21500 PUNE
21600 MUMBAI
21900 CHENAI
22000 HYDERABAD
//GO.DD3 DD SYSOUT=*
//LKED.SYSLMOD DD DSN=XX7384A.COBOL.LOADLIB(&MEM),DISP=SHR
//*
Output :
21500 MANDIP 44444 PUNE
21600 RANDIP 99999 MUMBAI
IDENTIFICATION DIVISION.
PROGRAM-ID.
FILCMPR.
ENVIRONMENT
DIVISION.
CONFIGURATION
SECTION.
OBJECT-COMPUTER.
IBM-390.
SOURCE-COMPUTER. IBM-390.
INPUT-OUTPUT
SECTION.
FILE-CONTROL.
SELECT
INFILE ASSIGN
TO DD1
FILE
STATUS IS FS1.
SELECT
OUTFILE
ASSIGN TO DD2
FILE
STATUS IS FS2.
SELECT
MCHFILE
ASSIGN TO DD3
FILE
STATUS IS FS3.
DATA
DIVISION.
FILE
SECTION.
FD
INFILE.
01
INREC.
05
EID1
PIC
X(05).
05
FILLER
PIC
X(01).
05
ENAME1
PIC X(06).
05
FILLER
PIC
X(01).
05
ESAL1
PIC
9(05).
05
FILLER
PIC
X(62).
FD
OUTFILE.
01
OUTREC.
05
EID2
PIC
X(05).
05
FILLER PIC X(01).
05
EADD2 PIC
X(10).
05
FILLER
PIC
X(64).
FD
MCHFILE.
01
MCHREC.
05
EIDO
PIC
X(05).
05
FILLER
PIC X(01).
05
ENAMEO PIC
X(06).
05
FILLER
PIC X(01).
05
ESALO
PIC
9(05).
05
FILLER
PIC X(01).
05
EADDO PIC
X(10).
05
FILLER
PIC X(51).
WORKING-STORAGE SECTION.
01
FS1
PIC X(02).
01
FS2 PIC X(02).
01
FS3
PIC
X(02).
01
SWITCH1 PIC X(01).
88
EOF1
VALUE
'Y'.
88
NOT-EOF1 VALUE 'N'.
01
SWITCH2 PIC
X(01).
88
EOF2
VALUE
'Y'.
88
NOT-EOF2 VALUE
'N'.
01
WS-EID
PIC
X(05).
PROCEDURE
DIVISION.
SET
NOT-EOF1 TO
TRUE
PERFORM
OPEN-PARA.
PERFORM READ-PARA UNTIL
EOF1.
PERFORM
CLOSE-PARA.
STOP RUN.
OPEN-PARA.
OPEN INPUT INFILE
OPEN OUTPUT MCHFILE.
DISPLAY ' INFILE
OPEN STATUS '
FS1.
DISPLAY ' MCHFILE OPEN
STATUS '
FS3.
READ-PARA.
IF FS1 =
'00'
READ
INFILE
AT
END
SET EOF1 TO
TRUE
NOT AT
END
MOVE EID1 TO
WS-EID
SET NOT-EOF2 TO
TRUE
PERFORM
OPEN-OUT-PARA
PERFORM READ-OUT-PARA
UNTIL
EOF2
PERFORM
CLOSE-OUT-PARA
END-READ
END-IF.
OPEN-OUT-PARA.
OPEN
INPUT OUTFILE
DISPLAY ' OUTFILE OPEN
STATUS '
FS2.
READ-OUT-PARA.
IF FS2 =
'00'
READ
OUTFILE
AT
END
SET EOF2 TO
TRUE
NOT
AT
END
IF
WS-EID =
EID2
MOVE EID1 TO
EIDO
MOVE ENAME1 TO
ENAMEO
MOVE ESAL1 TO ESALO
MOVE EADD2 TO EADDO
WRITE MCHREC
END-IF
END-READ.
CLOSE-OUT-PARA.
CLOSE
OUTFILE
DISPLAY ' OUTFILE CLOSE
STATUS '
FS2.
CLOSE-PARA.
CLOSE
INFILE
CLOSE
MCHFILE.
DISPLAY ' INFILE
CLOSE STATUS ' FS1.
DISPLAY ' MCHFILE CLOSE
STATUS '
FS3.
JCL to Execute above cobol program -
//STEP010 EXEC IGYWCLG,MEM=FILCMPR
//SYSPRINT DD SYSOUT=*
//COBOL.SYSIN DD DSN=XX7384A.JCLLIB.SELFTEST.Y2013.COB(&MEM),
// DISP=SHR
//GO.DD1 DD *
21400 SANDIP 33333
21500 MANDIP 44444
21600 RANDIP 99999
21700 BALDIP 66666
//GO.DD2 DD *
21500 PUNE
21600 MUMBAI
21900 CHENAI
22000 HYDERABAD
//GO.DD3 DD SYSOUT=*
//LKED.SYSLMOD DD DSN=XX7384A.COBOL.LOADLIB(&MEM),DISP=SHR
//*
Output :
21500 MANDIP 44444 PUNE
21600 RANDIP 99999 MUMBAI
Above program will degrade the performance so this is not advisable.
Option2 program is efficient.
Option2:
NOTE: Both the files are sorted on key and duplicates are removed from both input files.
IDENTIFICATION DIVISION.
PROGRAM-ID. COMPFIL.
ENVIRONMENT DIVISION.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
SELECT INFILE1 ASSIGN TO DD1.
SELECT INFILE2 ASSIGN TO DD2.
SELECT INFILEB ASSIGN TO DD3.
SELECT INFILE1O ASSIGN TO DD4.
SELECT INFILE2O ASSIGN TO DD5.
DATA DIVISION.
FILE SECTION.
FD INFILE1.
01 INREC1.
05 ID1 PIC 9(03).
05 FILLER PIC X(77).
FD INFILE2.
01 INREC2.
05 ID2 PIC 9(03).
05 FILLER PIC X(77).
FD INFILEB.
01 INRECB.
FILLER PIC X(80).
FD INFILE1O.
01 INREC1O.
FILLER PIC X(80).
FD INFILE2O.
01 INREC2O.
FILLER PIC X(80).
WORKING-STORAGE SECTION.
01 WS-VAR.
05 SWT1 PIC X(01).
88 EOF1 VALUE 'Y'.
88 NOT-EOF1 VALUE 'N'.
05 SWT2 PIC X(01).
88 EOF2 VALUE 'Y'.
88 NOT-EOF2 VALUE 'N'.
05 WS-ID1 PIC 9(03).
05 WS-ID2 PIC 9(03).
PROCEDURE DIVISION
SET NOT-EOF1 TO TRUE
SET NOT-EOF2 TO TRUE
OPEN INPUT INFILE1
INFILE2
OPEN OUTPUT INFILEB
INFILE1O
INFILE2O
PERFORM READ-FILE1
PERFORM READ-FILE2
PERFORM MTCH-PARA
PERFORM CLOS-PARA.
READ-FILE1.
READ INFILE1
AT END
SET EOF1 TO TRUE
NOT AT END
MOVE ID1 TO WS-ID1
END-READ.
READ-FILE2.
READ INFILE2
AT END
SET EOF2 TO TRUE
NOT AT END
MOVE ID2 TO WS-ID2
END-READ.
MTCH-PARA.
IF EOF1 AND EOF2
GO TO CLOS-PARA
END-IF
IF EOF1 AND NOT-EOF2
PERFORM READ-FILE2
INITIALIZE INREC2O
MOVE INREC2 TO INREC2O
WRITE INREC2O
GO TO MTCH-PARA
END-IF
IF NOT-EOF1 AND EOF2
PERFORM READ-FILE1
INITIALIZE INREC1O
MOVE INREC1 TO INREC1O
WRITE INREC1O
END-IF
IF WS-ID1 = WS-ID2
INITIALIZE INRECB
MOVE INREC1 TO INRECB
WRITE INRECB
PERFORM READ-FILE1
PERFORM READ-FILE2
GO TO MTCH-PARA
ELSE
IF WS-ID1 < WS-ID2
INITIALIZE INREC1
MOVE INREC1 TO INREC1O
WRITE INREC1O
PERFORM READ-FILE1
GO TO MTCH-PARA
ELSE
INITIALIZE INREC2O
MOVE INREC2 TO INREC2O
WRITE INREC2O
PERFORM READ-FILE2
GO TO MTCH-PARA
END-IF
END-IF.
CLOS-PARA.
CLOSE INFILE1
INFILE2
INFILEB
INFILE1O
INFILE2O.
STOP RUN.
JCL to run this program:
//AB7384A JOB (ACCT,U,A),'COMPFIL',CLASS=B,
// MSGCLASS=O,NOTIFY=&SYSUID
//*
// JCLLIB ORDER=AB7384A.SOURCE.PROCLIB
//STEP1 EXEC IGYWCLG,MEM=COMPFIL
//COBOL.SYSIN DD DSN=AB7384A.SOURCE.COBOL(&MEM),DISP=SHR
//GO.DD1 DD DSN=AB7384A.TEST.IN1,DISP=SHR
//GO.DD2 DD DSN=AB7384A.TEST.IN2,DISP=SHR
//GO.DD3 DD DSN=AB7384A.TEST.IN3,DISP=OLD
//GO.DD4 DD DSN=AB7384A.TEST.IN4,DISP=OLD
//GO.DD5 DD DSN=AB7384A.TEST.IN5,DISP=OLD
//LKED.SYSLMOD DD DSN=AB7384A.COBOL.LOADLIB(&MEM),DISP=SHR
//GO.SYSIN DD *
//*
Input File1:
080
100
125
150
175
200
Input File2:
020
040
080
090
095
100
150
160
170
175
200
210
220
Match file:
080
100
150
175
200
Nomch file1: (No Match Data from file1)
125
Nomch file2: (No Match Data from file2)
020
040
090
160
170
210
220
Don’t we have to sort the files before compare?
ReplyDeleteif it is sorting order then only above coding will work.
DeleteSimply superb... :)
ReplyDeleteWe can do same logic for two ksds files using index organization?
ReplyDeleteyes you can
Delete