Rewind a sequential file in COBOL program

kalais · New User Joined: 18 Apr 2006 Posts: 22

Hi,

I need to search a sequential file repeatedly each time from the start of the file.
How I can achieve this without closing and opening the file each time.
Is there any options like Rewind for doing it?

Thanks,
Kalais

dbzTHEdinosauer · Posted: Thu Aug 09, 2007 8:49 pm

No rewind for dasd. you could considered tabling the sequential file, thus reducing the I/O.

But, why do you have to make repeated passes? A redesign solution may be provided if you can provide your requirements.

jz1b0c · Posted: Thu Aug 09, 2007 8:49 pm

Sort the file on some key, (let there be duplicates- not an issue)

Now once you are at the end of the file, use START command to position it to the beginning of the file (use low-values in the key field).

I think you have to use access method dynamic for this.

Its been a long time, since I worked with files in cobol, I am not able to quote an example here, but I think this should work.

stodolas · Posted: Thu Aug 09, 2007 9:41 pm

Full file searches are generally not a good thing if they are done repeatedly in program. The cpu performance and run time of your application is going to suffer greatly by doing this.

dick scherrer · Posted: Thu Aug 09, 2007 11:09 pm

Hello,

If you explain why you need to search the file repeatedly, we may be able to offer far better performing alternatives.

Devzee · Posted: Fri Aug 10, 2007 10:31 am

See if you can use VSAM KSDS - this will avoid repeated scanning of the file.

kalais · New User Joined: 18 Apr 2006 Posts: 22

Hi,

I used a table for storing the sequntial file and did the search. The program got considerable improvement in performance.

Maximum no of records for that file will be below 5000 I hope.

Problem definition is that,
There are two sequential files FILE1 and FILE2 .
There will be a field in FILE1 with missing value. I need to search FILE2
using one or more key values(these keys for search will differ based on a field in FILE1) from FILE1 to find out the missing value.

(e.x)
FILE1
------
1.ACMABANO <missing field> PAP1 R PRDW ACMABANO
2.APFL905 <missing field> PAP1 L

If the field in BOLD is R ,search keys will be PRDW and ACMABANO
or if it is L search key will be the first field APFL905.

FILE2
------
1.ACMABANO PVSAM.CICS.ACMABANO PRDW L
2. APFL905 PVSAM.CICS.APFL905 PRDC L

The BOLD fields in FILE2 are the results of serach.

If anybody has better idea please let me know.

Please let me know if more clarification is needed.

Thanks,
Kalai