Question

Having the next table:

id -> incremental
field_1 -> foreignkey
field_2 -> foreignkey

I want to add the next index

ALTER TABLE my_table ADD unique index(field_1, field_2);

How ever I have (due a bad application validation) I have a lot of repeated rows (by repeated I mean same field_1 and same field_2, having just id as difference)

The table has about 60,000 rows so… removing field by field would be very hard.

How can I apply that index and remove every duplicated row?

Answer 1

create table mytable2 like mytable;

insert into mytable2 
select max(id), field_1, field_2 
from mytable 
group by field_1, field_2;

rename table mytable to mytable_old, mytable2 to mytable;

ALTER TABLE my_table ADD unique index(field_1, field_2)