Question

I entered a query that introduced some duplicates into my database. The table is straight forward.

It has an id (int) column and a phrase column which is varchar(255). In order to find duplicates, my query looks like the following:

SELECT phrase from foo GROUP BY phrase HAVING (count(phrase) > 1)

My question is, how do I delete the duplicate entries without manually having to do it? I want to use the query above to generate the list of entries that need to be deleted at least once. This way only one version of ‘phrase’ exists in table foo.

Answer 1

This would keep one row (the one with the lowest ID) per phrase.

DELETE FROM foo
WHERE id NOT IN (
    SELECT id FROM (
        SELECT MIN(id) id
        FROM foo
        GROUP BY phrase
    ) _
);

As dan pointed out in comments, with MySQL you need that weird inner query.