Querying two flat files on common key in COBOL

ganeshptrk · New User Joined: 04 Apr 2007 Posts: 17 Location: Chennai

Hi,
The following question was asked in one of the interviews.

Problem statement:
I have two files say FILE1 and FILE2

Contents of FILE1
EmpNo EmpName
1 Latha
2 Mahesh
3 Suresh
7 Ganesh
and so on... Approximately 10 million records in FILE1

Contents of FILE2
EmpNo Salary
2 30000
3 25000
7 18000
9 21000
and so on... Approximately 20 million records in FILE2

I want to get information from both the file where EmpNo is same and write FILE 3, for example
Here the matching EmpNos are 2, 3 and 7
EmpNo EmpName Salary
2 Mahesh 300000
3 Suresh 250000
7 Ganesh 180000

And non-matching records, for example 1 and 9 in above case, should be written to file 4. There should also be EOF handling routine.

Can anyone tell me the procedure division logic.

thanks
Ganesh

dick scherrer · Posted: Sun Jun 24, 2007 5:49 pm

Hello,

What you are asking about is a called 2-file match/merge) or a "line balance" bit of code. This has been discussed multiple times in the forums (primarliy in COBOL).

Use the search facility and when you find something that you don't understand, please let us know what is your doubt and we will be able to offer suggestions.

expat · Posted: Sun Jun 24, 2007 6:13 pm

Also search the SORT forum, as that offers an alternative solution without having to spend ages writing a program.

dick scherrer · Posted: Sun Jun 24, 2007 11:00 pm

Hello,

The cobol code shouldn't take any longer than the sort control statements - once there is a working model (which should already exist - IMHO).

Is it good to keep in mind that the only code that should need to be added to a decent model/skeleton is the actual fields for the match and any "other than match" logic required to meet the business requirement.