I have 2 files.The first file contains only one record,in each column of that record i would be having the custid field.
In the 2nd file i would be having the records for custid like custid,name,description
File 1
100 200 300 400
File 2
100 abc
200 ade
350 ref
800 retag
my output should be
100 abc
200 ad
There must be one to one match between the first column custid and the first row cust id,similarly the second column custid must be matched with the 2nd row custid
Only the matched records must be written to the output file
The number of columns in the first file is directly proportional to the number of rows in the second file
The custid field is of length 3 and they are delimited by spaces in the first file
I guess it can be done
but If my understand is right, if you have only custid's in first file and 2nd file holds custid's and other details..then your output should be custid's with details..then can't we use directly the second file?
If you want only the details which custid's present on first file yes, then does that mean the number of custid's are different to second & first file?
Joined: 30 Dec 2006 Posts: 20 Location: Kolkata, India
This might be more complicated than it looks. Most tools on mainframes work row-wise, that is one row at a time. Your problem is simplified many-fold if you can get the first dataset transposed, i.e. say it is like:
100
200
300
400
Then you can do a MERGE FEILDS and get your result.
But I am not aware of any tool which can directly transpose a dataset in such fashion. To come up with a solution using existing tools will be complicated, time consuming and would probably have a lot of limitations. Try using any other technology that is available with you to do this, I bet it will be simpler.
Joined: 23 Nov 2006 Posts: 19244 Location: Inside the Matrix
Hello,
How is file1 created? What happens if there are more custid's than can fit in one record?
If this is something new and is still being tested, it may be to your advantage to change how file1 is created and put only 1 custid per record rather than some series of them.
