DFSORT equivalent to Comparex job

Chirag Dhull · New User Joined: 29 Aug 2008 Posts: 8 Location: Bangalore

Hi !!

I have a Comparex Job which Compares PREVTERM Vs CURRTERM shows O/p like this :-

PREVTERM :

333444555 KRYG M E MARK E KRYG V1

CURRTERM :

333444555 KRYG M E MARK E KRYG V1

333444555 KRYG M E MARYLN F KRYG V6

333444555 KRYG M E MARK E KRYG V1

333444555 KRYG M E ELIGIBLE CHILDREN V8

Comparex OUTPUT :-

333444555 KRYG M E MARYLN F KRYG V6

333444555 KRYG M E MARK E KRYG V1

333444555 KRYG M E ELIGIBLE CHILDREN V8

can You tell me how to get the same O/P using DFSORT

Skolusu · Posted: Wed Sep 03, 2008 9:57 pm Post subject: Reply to: DFsort Problem

Chirag Dhull,

The following DFSORT/ICETOOL JCL will give you the desired results. I assumed that both your input files are FB recfm and 80 bytes in length. I also assumed that you have duplicates in both files and your comparison key is full 80 bytes. Make sure that you have Disp=MOD for the temp file T1.

Chirag Dhull · New User Joined: 29 Aug 2008 Posts: 8 Location: Bangalore

Hi ,

Actually i tried running this code , but i am not getting the O/p which i need .Is it because only one of my files has duplicate entries and other doesnt ?

My Comparex takes the Prev file and curr file as input with SYSIn :- KEY(1,9) , FIELD (1,9) .COPYDIFF , MLC=40.

CURR file has duplicate entries(Lrecl of files is 50)
PREV file doesnt have (Lrecl of files is 50 )

For e.g ;-
CURR File:-

333444555 KRYG M E 7 LANE RICHMOND STREET XYZ CITY
333444555 KRYG M E 7 LANE RICHMOND STREET XYZ CITY
445556677 BOB R W 9 LANE HARMISON STREET ABC CITY

PREV File;-

333444555 KRYG M E 7 LANE RICHMOND STREET XYZ CITY
445556677 BOB R W 9 LANE HARMISON STREET ABC CITY

O/p :-

333444555 KRYG M E 7 LANE RICHMOND STREET XYZ CITY

This is the O/p which Comparex gives me . Please tell me how to acheive this using some other tool .

Skolusu

Chirag Dhull,

The job shown above DOES give you the desired results. You just need to modify it according to your key comparison. The job shown above is comparing 80 bytes as key and now you only have 9 bytes to compare. You need to understand the job and modify it accordingly.

Anyway here is a job which will give you the desired results

Chirag Dhull · New User Joined: 29 Aug 2008 Posts: 8 Location: Bangalore

Let me try to explain it in detail :

I have file1 as:

key1
key1
Key2
key3
key3
key3
key5

and File2 as
key1
key1
key1
key2
key3
key3
key4

and the outut shd have :

key1
key4

which means from file2 whatever is not found in file 1 should be there in the output.

Like in case above we dont have key4 in file 1 and also we have one more ocuurence of key 1 in file1. So the output contains both of them...

Skolusu · Posted: Mon Sep 15, 2008 8:59 pm Post subject:

Chirag Dhull,

The following DFSORT/ICETOOL JCL will give you the desired results

Chirag Dhull · New User Joined: 29 Aug 2008 Posts: 8 Location: Bangalore

Thanks Kolusu for the reply...

But , the solution given above works till some extent in my case but its still not giving the desired result.

The problem is that in some of the records few fields gets changed(ie post marriage surname and all ). So overlay doesnt work properly if taken entirely.

I need to have it like restart(1,9) and then restart(40,10) to have the sequence number properly. assuming we have surname in between 10-39 byte.

Please suggest on how to achieve this.

Skolusu

Chirag Dhull,

*sigh* you need to spell out your requirements properly. I think the following DFSORT/ICETOOL job will give you the desired results based on your description. if it doesn't work then you need to show me an example of your input and desired output with the DCB properties and also explain the rules in detail