Question

I want to remove duplicate records from a table without using temp table how is it possible?

Answer 1

1. Make a query A that gets you the rows that qualify as duplicates
2. Make a query B that gets you the rows that you want to keep
3. Make a query that deletes all rows that are in A but not in B.

Example.

Assuming a table named table, with an autoincrement ID column named id and a column named name of which you want to remove the doubles. Of each name, you want to keep the oldest record (the one with the lowest ID).

Query A then looks something like:

SELECT * FROM table 
WHERE name IN (SELECT name FROM table GROUP BY name HAVING COUNT(*) > 1)

Query B would be:

SELECT * FROM table 
WHERE id IN (SELECT min(id) FROM table GROUP BY name)

Now combine these to form the delete query:

DELETE FROM table 
WHERE name IN (SELECT name FROM table GROUP BY name HAVING COUNT(*) > 1)
AND NOT id IN (SELECT min(id) FROM table GROUP BY name)

In the example at hand, you could leave out the first query, but when things get more complicated, it’s a nice extra safeguard.