Hello Thom.
what is the meaning of
select table_name from table_name
Also is this a common behavior of all Databases i.e. oracle , Microsoft ,...etc . i.e is this is the standard
behavior
I think this is a good way to find duplicates in general, I will write a routine to compare all the columns by excluding the primary key which is serial
Thanks in advance
Regards
From: Thom Brown <thom@xxxxxxxxx>
To: "Gauthier, Dave" <dave.gauthier@xxxxxxxxx>
Cc: Andy Colson <andy@xxxxxxxxxxxxxxx>; "pgsql-general@xxxxxxxxxxxxxx" <pgsql-general@xxxxxxxxxxxxxx>
Sent: Wednesday, September 7, 2011 12:50 AM
Subject: Re: Deleting one of 2 identical records
On 6 September 2011 19:00, Gauthier, Dave <dave.gauthier@xxxxxxxxx> wrote:
The identification and deleting of the records using ctids seems to have worked fine.
Thanks !
Alternative you could do something like this:
WITH keep AS (
SELECT
my_table AS duplicate_row,
min(ctid) AS keep,
count(*)
FROM my_table
GROUP BY my_table
HAVING count(*) > 1
)
DELETE FROM my_table
USING keep
WHERE
my_table = keep.duplicate_row
AND
my_table.ctid != keep
RETURNING my_table.ctid, my_table.*;
This would delete all duplicate rows from the table and just keep whichever row appears first in the table before its duplicates.
Thom Brown
Twitter: @darkixion
IRC (freenode): dark_ixion
Registered Linux user: #516935
EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
