How can I join two tables with different number of rows in MySQL?

Question

I have two tables which I want to connect.

TABLE_A:

+-----------+-----------+---------+
| row_id    | category  | val_1   |
+-----------+-----------+---------+
| 1067      | cat1      | 6.5     |
| 2666      | test      | 6.5     |
| 2710      | cat1      | 2.1     |
| 2710      | test      | 7.1     |
| 2767      | test      | 3       |
| 71142     | cat1      | 5       |
| 50666     | other     | 6.5     |
| 71142     | other     | 1       |
| 345342    | cat1      | 6.5     |
| 345342    | test      | 2.8     |
+-----------+-----------+---------+

TABLE_B:

+-----------+-----------+
| row_id    | val_2     |
+-----------+-----------+
| 1067      | 2.0       |
| 2666      | 9         |
| 2701      | 2.2       |
| 2708      | 1         |
| 2709      | 6.5       |
| 2710      | 5.2       |
| 2765      | 6.5       |
| 2766      | 15        |
| 2767      | 8         |
| 71142     | 5         |
| 2783      | 4.5       |
| 50666     | 6.5       |
| 101588    | 9         |
| 101588    | 3         |
| 3452      | 8.0       |
| 23422     | 5         |
| 345342    | 6.5       |
+-----------+-----------+

RESULT_TABLE:

+-----------+-----------+-----------+------------+
| row_id    | val_2     | val_1     | category   |
+-----------+-----------+-----------+------------+
| 1067      | 2.0       | 6.5       | cat1       |
| 2666      | 9         | 6.5       | test       |
| 2701      | 2.2       | 2.2       | NULL       |
| 2708      | 1         | 1         | NULL       |
| 2709      | 6.5       | 1         | NULL       |
| 2710      | 5.2       | 2.1       | cat1       |
| 2710      | 5.2       | 7.1       | test       |
| 2765      | 6.5       | 1         | NULL       |
| 2766      | 15        | 1         | NULL       |
| 2767      | 8         | 3         | test       |
| 71142     | 5         | 5         | cat1       |
| 71142     | 5         | 1         | other      |
| 2783      | 4.5       | 1         | NULL       |
| 50666     | 6.5       | 6.5       | other      |
| 101588    | 9         | 1         | NULL       |
| 101588    | 3         | 1         | NULL       |
| 3452      | 8.0       | 1         | NULL       |
| 23422     | 5         | 1         | NULL       |
| 345342    | 6.5       | 6.5       | cat1       |
| 345342    | 6.5       | 2.8       | test       |
+-----------+-----------+-----------+------------+

I tried to use something like this:

SELECT TABLE_A.row_id, TABLE_A.category, TABLE_A.val_1, TABLE_B.val_2
FROM TABLE_A
INNER JOIN TABLE_B ON TABLE_B.row_id = TABLE_A.row_id
ORDER BY row_id;

However, the result included only rows where the row_id column exists in TABLE_A.

Is there way to connect TABLE_A and TABLE_B to produce the result shown in RESULT_TABLE?

Solution 1

If you want all the results, you need an outer join, not an inner one. (Inner only returns the rows where there is a match; outer returns all rows, with the matching rows "stitched together")

Solution 2

Try an outer join.

SELECT TABLE_A.row_id, TABLE_A.category, TABLE_A.val_1, TABLE_B.val_2
FROM TABLE_B
LEFT OUTER JOIN TABLE_A ON TABLE_B.row_id = TABLE_A.row_id
ORDER BY row_id;

Solution 3

This ought to do it:

SELECT 
  TABLE_B.row_id row_id, 
  TABLE_A.category category, 
  COALESCE(TABLE_A.val_1,1) val_1,
  TABLE_B.val_2 val_2
FROM TABLE_A
RIGHT OUTER JOIN TABLE_B 
  ON TABLE_B.row_id = TABLE_A.row_id
ORDER BY TABLE_B.row_id;

The RIGHT OUTER JOIN pulls all the records from Table_B even if they don't exist in Table_A, and the COALESCE statement is a function that returns its first non-NULL parameter. In this case, if there is no value in Table_A, it will return 1, which is what your example result lists as the desired output.

Comments

[Michael Todd]

Try a full join instead.