Question

Assuming three tables which map a many to many relationship.

People
1 Bob
2 Mark
3 Peter
4 Tracy
5 Robert
6 ...

Favorite_Food
1 Icecream
2 Fish&Chips
3 Chocolate
4 Pizza

Pople_To_Favorite_Food
1, 2 # Bob likes Fish&Chips.
2, 2 # Mark likes Fish&Chips too.
4, 2
3, 3
3, 2
1, 1 # Bob likes Icevream.
3, 1
4, 1
5, 3

What is the recommended way to realize the following query:

1. Selecting every favorite food of Bob.
2. Then select every person, who shares at least one similar favorite food with Bob.
3. List the names found.

So what I am looking for is some kind of back reference.

In this example it would have to result in: Bob likes Icecream and Fish&Chipy; Mark, Peter and Tracy like at least one food which Bob likes. So Mark, Peter and Tracy are listed.

I thought of doing something like

SELECT fk_people_id FROM [...join the tables here...] WHERE favorit_food_id IN (
SELECT fk_favorite_food_id [...join the tables here...])

Is this the way to go, or is there a more suitable concept I should be using?

Perhaps I should mention, that the table I want to query is really huge (Millions of entrys).

Answer 1

Try this one –

SELECT p2.* FROM People p1
  JOIN Pople_To_Favorite_Food pf1
    ON pf1.people_id = p1.people_id
  JOIN Favorite_Food f1
    ON f1.food_id = pf1.food_id
  JOIN Pople_To_Favorite_Food pf2
    ON pf2.food_id = f1.food_id
  JOIN People p2
    ON p2.people_id = pf2.people_id
WHERE
  p1.name = 'Bob'
GROUP BY
  p2.people_id;