File processing with Maximun count as a field

dharmendra.rawat84 · New User Joined: 26 Oct 2010 Posts: 1 Location: Hyderabad

Hi,

I have a file having records like this mentioned below.

SYS-ID PROV-ID COUNT

111111 3333333 10
111111 4444444 20
111111 5555555 15
222222 6666666 10
222222 7777777 20
222222 8888888 15

I have a requirement like I want the out file to be as below.
I want the records in output file that has max count.

SYS-ID PROV-ID COUNT

111111 4444444 20
222222 7777777 20

Please give me some suggestions...

Regards,

Dharam

dbzTHEdinosauer · Posted: Wed Dec 01, 2010 9:17 pm

if you know what the value of the max-count is, no problem
otherwise, sort the file prior to reading (max-count Desc).

Kjeld · Posted: Thu Dec 02, 2010 7:13 pm

Or to put it more complete:

Sort on keys ASCENDING SYS-ID PROV-ID, DESCENDING COUNT and subsequently process the first record in each break of the key SYS-ID PROV-ID, and discard the rest.

dbzTHEdinosauer · Posted: Thu Dec 02, 2010 8:24 pm

actually, you definition of the problem is confusing.

what are the actual control breaks?

is the highest count the highest of the file?
will every sys-id have a record (regardless of prov-id)
that contains this high-count.
or do you want the sys-id/prov-id/count for every sys-id?