RE: Best Way to find duplicate records in same file

[Mark Allen <mallen@xxxxxxxx>]

Thanks all, went with keyed file approach (only 2 fields in file...210
bytes long)

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Goodbar Loyd ETS -
Water Valley
Sent: Friday, May 09, 2003 2:57 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: Best Way to find duplicate records in same file


Do you need to check the entire record, or a "key field"?

If it's just a key field, something like this SQL

SELECT DISTINCT(keyfield), COUNT(keyfield)
FROM somefile

May work. If you need to do the entire record (is it one field or
multiple fields?), you can do the same but it looks like

SELECT DISTINCT(field1 || field2 || char(field3) || ...), count(*) /* I
think */ FROM somefile

HTH,
Loyd

--  
Loyd Goodbar
Programmer/analyst
BorgWarner Incorporated
ETS/Water Valley
662-473-5713
lgoodbar@xxxxxxxxxxxxxx


-----Original Message-----
From: Mark Allen [mailto:mallen@xxxxxxxx] 
Sent: Friday, May 09, 2003 12:16 PM
To: midrange-l@xxxxxxxxxxxx
Subject: Best Way to find duplicate records in same file


I need to add duplicate record checking in a file we received from a
switch. (I'll just add it to the CL that already FIXES a bunch of stuff
in the file
anyways) and was wondering what the best (i.e easy/fast) to do it would
be. I've come up with:
 
1.  Run a query against the file, sorted by entire record, level break
on entire record, count.  Output Query results to file and then do cpyf.
2. Build a file keyed on all 210 bytes, read original chain to new file,
if not found write it 3.  SQL ???  not done much with it but maybe it
could do? _______________________________________________
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe,
unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo.cgi/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.