Merging two files into one based on certain key fields:

[MVS_blob]

Hi,

I've got a tricky one for those up for a challenge.

I have two input files, and want to create an output file. I'm adding a two digit number to a file containing a header and detail records. Note: the two digit number is dependent on the header and detail records.

INPUT1:

[MVS_blob]

Been working on this problem today, and have gotten somewhere with it.

My steps were: (excuse my noob-ness)

1. program: DFSORT
-extract all HEADER rows to FILEA & append a 6 digit sequence number to the end
-extract all DETAIL rows to FILEB & append a 6 digit sequence number to the end
2. program: DFSORT
- remove duplicates in FILEB (DEATIL rows)
3. program: DFSORT
- sort FILEB according to the 6 digit sequence number.
- re-number the 6 digit sequence number append to FILEA
4. program: ICETOOL
- SPLICE the FILEA to get FILEC which looks like

[Frank Yaeger]

Yes, this is tricky.

Here's a DFSORT/ICETOOL job that will do what I think you asked for. I assumed your input files have RECFM=FB and LRECL=80, but the job can be changed appropriately for other attributes.

I also assumed that:
- positions 5-14 of input file1 are to be matched with positions 7-16 of the HEADER record in input file2, AND
- positions 1-3 of input file1 are to be matched with positions 8-10 of the first DEATIL record after the HEADER record in input file2

[MVS_blob]

Frank, you're a genius!!!

Thanks soo much! It works perfectly.

My original code was 100+ lines, was slow, and only did 1/3 of the things yours does.

Great work!

[Frank Yaeger]

Thanks for the feedback and the kind words. Glad I could help.
_________________
Frank Yaeger - DFSORT Development Team (IBM) - yaeger@us.ibm.com
Specialties: JOINKEYS, FINDREP, WHEN=GROUP, ICETOOL, Symbols, Migration
=> DFSORT/MVS is on the Web at http://www.ibm.com/storage/dfsort/