How to delete a duplicate row in a flat file

ousep143 · New User Joined: 06 Oct 2007 Posts: 32 Location: India

Hi,
How to delete a duplicate row in a flat file?Is it possible?

Thanks & Regards

[/b]

acevedo · Active User Joined: 11 May 2005 Posts: 344 Location: Spain

you should create a new file with those records omitted.

raak · Posted: Thu Jan 17, 2008 5:43 pm

If u checked the forums, u might have found out that u can use a simple sort card to achieve this...

expat · Posted: Thu Jan 17, 2008 5:48 pm

It means keep the original input and create a new output, do not over write the original input file.

ousep143 · New User Joined: 06 Oct 2007 Posts: 32 Location: India

Hi raak,

I want to know,how to delete duplicate record in a flat file or how to compare two records?

Regards

enrico-sorichetti · Posted: Thu Jan 17, 2008 5:52 pm

Acevedo' s reply simply meant that for flat (wrong term anyway) aka PS files
it is not possible to delete/insert records in place

do not confuse things when in the Ispf environment,... when editing a PS file
ISPF will just bring all the records in memory and afterwards ( at save time )
it will open the dataset in output and write out all the records

ousep143 · New User Joined: 06 Oct 2007 Posts: 32 Location: India

hi enrico,

Is it possible to findout duplicate rows in a PS file.

Regards

acevedo · Active User Joined: 11 May 2005 Posts: 344 Location: Spain

expat and enricho assumptions are correct, of course.

ousep143
1-read the posts about delete a PS file.
2-compare... that's another question.

this post is in the Cobol forum...maybe you shoud try a search in the DFSORT/ICETOOL forum.

;)

raak · Posted: Thu Jan 17, 2008 6:03 pm

Just curious...

the original question was

acevedo · Active User Joined: 11 May 2005 Posts: 344 Location: Spain

IMNHO... we should know if this has to be done in Cobol or not.

IBMMAINFRAMES.com Support Forums -> Mainframe COBOL

enrico-sorichetti · Posted: Thu Jan 17, 2008 6:14 pm

ousep143 · New User Joined: 06 Oct 2007 Posts: 32 Location: India

Hi everybody,
Thanks for information.

Regards

mnnarayankar · New User Joined: 21 Jan 2008 Posts: 23 Location: bangalore

nice questions.

One answer is this 1st load u r PS file into KSDS vsam dataset. using COBOL pgm or using REPRO command. VSAM KSDS allow to write only unique records with help of key. Once all unique data u get then unload VSAM file into PS file.

mnnarayankar · New User Joined: 21 Jan 2008 Posts: 23 Location: bangalore

use SORT utilities to elinminates the duplicate within a file

Syntax : Sort fields = (starting pos,length,type,asc/desc),EQUALS
Sum fields = none.

use INDD, OUTDD, and SORTWORK

starting position : anything but integer value.
length : Length mean total no of character from 1st character.
type means : ZD,PD,BI,CH

if EQUALS is coded then records to be written is the first record else it can be anything..