Selecting Records based on a condition

[aravind83]

Hi,

Need help in DFSORT.

I/P File :- RECFM=VB and LRECL=5230. The following is the sample input file
col1 is an 10 char fixed length. This will be randomly generated.
col2 is a single char fixed length. The values for this field are either 1 or 2.

col1********* col2******
123456abcd 1
123456abcd 1
123456abcd 2
123456acde 1
123456acde 1
123456acde 1
123456acde 1
123456acde 1
123456acde 1
123789abcf 1
123789abcf 1
123789abcf 1
123789abcf 1
123789abcf 2

Now i am trying to generate a o/p file as suggested below. The condition is that i would need all those records who have both 1 and 2 in col2 in the file i.e.

col1********* col2******
123456abcd 1
123456abcd 1
123456abcd 2
123789abcf 1
123789abcf 1
123789abcf 1
123789abcf 1
123789abcf 2

The file paramters will remain the same i.e RECFM=VB and LRECL=5230

[skolusu]

Aravind,

The following DFSORT JCL will give you the desired results. Since your input is VB , col1 actually starts at 14 (10 + 4 byte RDW) and col2 starts at pos 25

Code: Select all

//STEP0100 EXEC PGM=SORT                                   
//SYSOUT   DD SYSOUT=*                                     
//SORTIN   DD DSN=your input vb 5230 file,DISP=SHR
//SORTOUT  DD DSN=&&T1,DISP=(,PASS),SPACE=(CYL,(X,Y),RLSE) 
//SYSIN    DD *                                             
  SORT FIELDS=COPY                                         
  INCLUDE COND=(25,1,CH,EQ,C'2')                           
  OUTREC OVERLAY=(25:C'0')                                 
/*                                                         
//STEP0200 EXEC PGM=SORT                                   
//SYSOUT   DD SYSOUT=*                                     
//SORTIN   DD DSN=&&T1,DISP=SHR                             
//         DD DSN=your input vb 5230 file,DISP=SHR
//SORTOUT  DD SYSOUT=*                                     
//SYSIN    DD *                                             
  SORT FIELDS=(14,10,CH,A,25,1,CH,A),EQUALS                 
  OUTREC IFTHEN=(WHEN=INIT,                                 
  BUILD=(1,4,25,1,X,SEQNUM,8,ZD,RESTART=(14,10),5)),       
  IFTHEN=(WHEN=GROUP,BEGIN=(7,8,ZD,EQ,1),PUSH=(6:5,1))     
  OUTFIL INCLUDE=(5,2,SS,EQ,C'10,20'),BUILD=(1,4,15)       
/*   

[aravind83]

I will try it out..Thanq very much for the reply Skolusu

[aravind83]

Hi Skolusu,

I have run the sort codes as suggested with a slight tweak to adjust the positions and is working perfectly fine. Thanq very very much. But would another small help from you i.e. can u please give me an high level overview of the second sort stmts i.e. for step STEP0200 so as to understand as to how this works.

Thankq once again,
Aravind

[skolusu]

aravind83 ,

Step0100 will create a temp file with all the '2' records and put a value of '0' at pos 25

Code: Select all

----+----1----+----2----+
             123456ABCD 0
             123789ABCF 0


step0200 will concatenate the temp file and original input and sort on the key fields and this how it looks like before OUTREC

Code: Select all

----+----1----+----2----+
             123456ABCD 0
             123456ABCD 1
             123456ABCD 1
             123456ABCD 2

             123456ACDE 1
             123456ACDE 1
             123456ACDE 1
             123456ACDE 1
             123456ACDE 1
             123456ACDE 1

             123789ABCF 0
             123789ABCF 1
             123789ABCF 1
             123789ABCF 1
             123789ABCF 1
             123789ABCF 2


OUTREC IFTHEN=(WHEN=INIT,
BUILD=(1,4,25,1,X,SEQNUM,8,ZD,RESTART=(14,10),5)),

This will put the indicator in pos 5 and also a seqnum at pos 7 whenever the key(14,10) changes

Code: Select all

    0 00000001 123456ABCD 0
    1 00000002 123456ABCD 1
    1 00000003 123456ABCD 1
    2 00000004 123456ABCD 2
    1 00000001 123456ACDE 1
    1 00000002 123456ACDE 1
    1 00000003 123456ACDE 1
    1 00000004 123456ACDE 1
    1 00000005 123456ACDE 1
    1 00000006 123456ACDE 1
    0 00000001 123789ABCF 0
    1 00000002 123789ABCF 1
    1 00000003 123789ABCF 1
    1 00000004 123789ABCF 1
    1 00000005 123789ABCF 1
    2 00000006 123789ABCF 2



IFTHEN=(WHEN=GROUP,BEGIN=(7,8,ZD,EQ,1),PUSH=(6:5,1))
OUTFIL INCLUDE=(5,2,SS,EQ,C'10,20'),BUILD=(1,4,15)

now if the seqnum we have at pos 7 we push down the value at 5 on pos 6 . now the contents will look like this

Code: Select all

00 00000001 123456ABCD 0 
10 00000002 123456ABCD 1 
10 00000003 123456ABCD 1 
20 00000004 123456ABCD 2 
11 00000001 123456ACDE 1 
11 00000002 123456ACDE 1 
11 00000003 123456ACDE 1 
11 00000004 123456ACDE 1 
11 00000005 123456ACDE 1 
11 00000006 123456ACDE 1 
00 00000001 123789ABCF 0 
10 00000002 123789ABCF 1 
10 00000003 123789ABCF 1 
10 00000004 123789ABCF 1 
10 00000005 123789ABCF 1 
20 00000006 123789ABCF 2 


Using OUTFIL include we only pick the values 10 and 20 and remove the unwanted fields we created.

I suggest that you read the User Guide for DFSORT PTF UK90013 paper (sortugpf.pdf) which explains in detail about the GROUP and other new functions

http://www.ibm.com/support/docview.wss? ... g3T7000085