CROSS JOIN vs INNER JOIN in SQL

Question

What is the difference between CROSS JOIN and INNER JOIN?

CROSS JOIN:

SELECT 
    Movies.CustomerID, Movie         


        

Answer 1

CROSS JOIN

As I explained in this article, the CROSS JOIN is meant to generate a Cartesian Product.

A Cartesian Product takes two sets A and B and generates all possible permutations of pair records from two given sets of data.

For instance, assuming you have the following ranks and suits database tables:

And the ranks has the following rows:

| name  | symbol | rank_value |
|-------|--------|------------|
| Ace   | A      | 14         |
| King  | K      | 13         |
| Queen | Q      | 12         |
| Jack  | J      | 11         |
| Ten   | 10     | 10         |
| Nine  | 9      |  9         |

While the suits table contains the following records:

| name    | symbol |
|---------|--------|
| Club    | ♣      |
| Diamond | ♦      |
| Heart   | ♥      |
| Spade   | ♠      |

As CROSS JOIN query like the following one:

SELECT
   r.symbol AS card_rank,
   s.symbol AS card_suit
FROM
   ranks r
CROSS JOIN
   suits s

will generate all possible permutations of ranks and suites pairs:

| card_rank | card_suit |
|-----------|-----------|
| A         | ♣         |
| A         | ♦         |
| A         | ♥         |
| A         | ♠         |
| K         | ♣         |
| K         | ♦         |
| K         | ♥         |
| K         | ♠         |
| Q         | ♣         |
| Q         | ♦         |
| Q         | ♥         |
| Q         | ♠         |
| J         | ♣         |
| J         | ♦         |
| J         | ♥         |
| J         | ♠         |
| 10        | ♣         |
| 10        | ♦         |
| 10        | ♥         |
| 10        | ♠         |
| 9         | ♣         |
| 9         | ♦         |
| 9         | ♥         |
| 9         | ♠         |

INNER JOIN

On the other hand, INNER JOIN does not return the Cartesian Product of the two joining data sets.

Instead, the INNER JOIN takes all elements from the left-side table and matches them against the records on the right-side table so that:

• if no record is matched on the right-side table, the left-side row is filtered out from the result set
• for any matching record on the right-side table, the left-side row is repeated as if there was a Cartesian Product between that record and all its associated child records on the right-side table.

For instance, assuming we have a one-to-many table relationship between a parent post and a child post_comment tables that look as follows:

Now, if the post table has the following records:

| id | title     |
|----|-----------|
| 1  | Java      |
| 2  | Hibernate |
| 3  | JPA       |

and the post_comments table has these rows:

| id | review    | post_id |
|----|-----------|---------|
| 1  | Good      | 1       |
| 2  | Excellent | 1       |
| 3  | Awesome   | 2       |

An INNER JOIN query like the following one:

SELECT
   p.id AS post_id,
   p.title AS post_title,
   pc.review  AS review
FROM post p
INNER JOIN post_comment pc ON pc.post_id = p.id

is going to include all post records along with all their associated post_comments:

| post_id | post_title | review    |
|---------|------------|-----------|
| 1       | Java       | Good      |
| 1       | Java       | Excellent |
| 2       | Hibernate  | Awesome   |

Basically, you can think of the INNER JOIN as a filtered CROSS JOIN where only the matching records are kept in the final result set.

For more details about how INNER JOIN works, check out this article as well.