|
View previous topic :: View next topic
|
| Author |
Message |
siva102
New User
Joined: 28 Nov 2007
Posts: 63
Location: Chennai
|
|
|
|
Hi All,
I am looking for some help in sort to merge two files and the results to capture in an output file.
Input File 1 : FB, 200 LRECL
Input File 2: FBA, 166 LRECL
File 1 Format:
| Code: |
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
XXXXX 02/16/201602/17/201602/18/2016
|
File 2 Format:
| Code: |
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
*** (AAA) 9,999.99 8,88.00 77,777.07
gjb jkhkjbn jhbgjk
ghfhjgfn jhygjhbjgh hjgjh
hngjk hjvjhbb
*** (WWW) 999,999.99 8,888.00 777,777.77
asdjkashdkjh lkajsdklj
asdbhkjjkl nhjklanmsd,n j
ajksdnhjkn lkkjjl
*** (ZZZ) 9,999.99 8,88.00 555,777.77
asda as sf asdad
*** (MMM) 9,909.99 8,88.00 555,777.77 |
The output should be as below:
| Code: |
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
Sample report for system
WWW 999,999.99hard core 2016021720160218values WWW
ZZZ 9,999.99hard code 2016021720160218values ZZZ |
Condition to create:
1) It will have the header records written in the file as "Sample report for system".
2) It will look in file-2, and pull the values from it only when the value in position 10 to 14 are either (WWW) or (ZZZ).
3) For both the scenario:first three bytes as www or zzz.
4) Starting from 5th byte, the amount field will be pulled from file 2 position 40 to 50.
5) Constant "hard core " value will be populated from 15 to 25.
6) for date it will look into the file 1. for position 26 to 33 it will use the value from 21 to 30 from file 1. In file 1 the dates are in MM/DD/YYYY but in the output file it should be in YYYYMMDD.
7) for position 34 to 41, it will use the file 1 from 31 to 40.
8) Again hard code values needs to be populated.
Can someone please help me in this regard. |
|
| Back to top |
|
 |
Lakshmi Prakash
New User
Joined: 22 Feb 2016
Posts: 5
Location: India
|
|
|
|
Hello,
On the first glance I could think of this approach
1) Remove unnecessary records from File 2
2) Add some dummy keys in front of file 1 and 2
3) Use Joinkeys and concatenate both File1 and file 2
4) Now you have both file 1 and file 2 in one line, you can reformat the line in the way you desire.
Not sure if this is any other better approach. |
|
| Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011
Posts: 7309
Location: Inside the Matrix
|
|
|
|
Lakshmi Prakash,
That approach would work, but JOINKEYS uses more resources.
Better would be a two-step approach. First step generates a DFSORT symbol/SMNAMES for the static data from the single-record file.
Then it is the INCLUDE COND=(what you want), do the necessary formatting, using the symbol where required. Use OUTFIL with REMOVECC and HEADER1 to get the heading.
Several examples here. |
|
| Back to top |
|
|
Lakshmi Prakash
New User
Joined: 22 Feb 2016
Posts: 5
Location: India
|
|
|
|
| I completely agree with your approach Bill, that's an efficient way. |
|
| Back to top |
|
|
|
|
  |
All times are GMT + 6 Hours |
|
