Question

When I have about 30 actors this query is taking 20+ seconds. Is there a way to speed this up dramatically? I am sure there is, I just don’t know in what way or where to start.

  REPEAT
    FETCH actors INTO a;
      IF a != '' THEN
        IF !(SELECT COUNT(*) FROM movieactor WHERE actor = a) THEN
          INSERT INTO movieactor (actor) 
          VALUES (a);
         END IF;
      END IF;
  UNTIL done END REPEAT;

movieactor table has just an actor and an id.

Answer 1

You really need to show the table definitions, indexes, query plan, and triggers to understand performance. However, this query has some obvious suggestions.

Why not use set operations, such as:

insert into movieactor(actor)
    select a
    from actors
    where a <> '' and
          a not in (select * from (select actor from movieactor))

(The double select is something that I think is needed for MySQL to parse this.)

Second, you should have an index on movieactor.actor. That will probably speed up the query a lot.