Deleting duplicate rows using sql query

[Eshwar Prasad]

I have a requirement to delete rows using sql query in oracle database. Actually for instance i have employee table having empname and emplocation.

When i write the sql statement as "select * from employee" i get the below results

empname emplocation

user1 india
user2 india
user1 india
user1 india
user3 india
user1 india
user2 india

i have a requirement to delete multiple occurrence of user1 and user2 and just retain one copy of the record.

so my final result set should have

user1 india
user2 india
user3 india

please let me know how can i do this. i tried using distinct, count etc but not meeting the requirement as above.

Your suggestions would be precious

[Srikanth Nalam]

Please mention the exact query you have used.
If you use select distinct empname, emplocation from employee you will get unique records with unique combination of these columns.

[Campbell Ritchie]

Welcome to the Ranch , Srikanth Nalam

[Srikanth Nalam]

Thanks Campbell

[Jan Cumps]

Srikanth,

this query might give you a starting point to remove the duplicates. Oracle has a concept called rowid. It gives all your records a unique identifier.
This rowid should typically not be used in any query (it is a relational database after all, we should never address a record this way), but it can help you in this occasion.
You have to find all records that have the same values in all columns (duplicates), and only keep one of them.
In my example, I have kept the one with the lowest rowid.

delete from <yourtable>
where
  rowid not in
  (
    select min(rowid)
    from <yourtable>
    group by <all columns in yourtable>
  )

Ensure that you understand the concept before trying it.
Test first on a test database!
Make a backup!
Etc..!


Regards, Jan