| Subject | Re: [firebird-support] How do find duplicates in a table? |
|---|---|
| Author | Alexey Kovyazin |
| Post date | 2016-02-04T19:56:49Z |
It is faster to use RDB$DB_KEY to find and delete duplicates.
See more details here
http://ib-aid.com/en/articles/how-to-find-and-delete-duplicate-records-in-firebird-database/
Regards,
Alexey Kovyazin
IBSurgeon
I had finally figured it out just before your reply
SELECT DISTINCT P.SOC_SEC_NO,
(SELECT COUNT(*)
FROM PERSON P2
WHERE P2.SOC_SEC_NO = P.SOC_SEC_NO) AS CNT
FROM PERSON P
WHERE P.SOC_SEC_NO IS NOT NULL
GROUP BY 1
HAVING (SELECT COUNT(*)
FROM PERSON P3
WHERE P3.SOC_SEC_NO = P.SOC_SEC_NO) > 1
This appears to work. See anything that I should change?
-----Original Message-----
From: firebird-support@yahoogroups.com
[mailto:firebird-support@yahoogroups.com]
Sent: Thursday, February 04, 2016 1:25 PM
To: firebird-support@yahoogroups.com
Subject: Re: [firebird-support] How do find duplicates in a table?
04.02.2016 20:09, 'stwizard' stwizard@... [firebird-support] wrote:
> How do I form a SQL Select statement that will return which records in
> my PERSON table have duplicate SOC_SEC_NO.
RTFM GROUP BY, HAVING, COUNT().
--
WBR, SD.