Select rows that are indirectly referenced in another table - mysql

Question

PREPARATION
Consider this script to create a MySQL dummy-database:
CREATE SCHEMA `zzz_dummy` ;
CREATE TABLE `zzz_dummy`.`subtable2` (
`id` INT NOT NULL UNIQUE,
`col1` VARCHAR(45) NULL,
PRIMARY KEY (`id`));
CREATE TABLE `zzz_dummy`.`subtable1` (
`id` INT NOT NULL UNIQUE,
`ref_subtab2` INT NULL,
PRIMARY KEY (`id`));
CREATE TABLE `zzz_dummy`.`maintable` (
`id` INT NOT NULL UNIQUE,
`ref_subtab1` INT NULL,
PRIMARY KEY (`id`));
ALTER TABLE `zzz_dummy`.`maintable`
ADD INDEX `fk_subtab1_idx` (`ref_subtab1` ASC);
ALTER TABLE `zzz_dummy`.`maintable`
ADD CONSTRAINT `fk_subtab1`
FOREIGN KEY (`ref_subtab1`)
REFERENCES `zzz_dummy`.`subtable1` (`id`)
ON DELETE NO ACTION
ON UPDATE NO ACTION;
ALTER TABLE `zzz_dummy`.`subtable1`
ADD INDEX `fk_subtab2_idx` (`ref_subtab2` ASC);
ALTER TABLE `zzz_dummy`.`subtable1`
ADD CONSTRAINT `fk_subtab2`
FOREIGN KEY (`ref_subtab2`)
REFERENCES `zzz_dummy`.`subtable2` (`id`)
ON DELETE NO ACTION
ON UPDATE NO ACTION;
INSERT INTO zzz_dummy.subtable2 VALUES
(1,'ref_val_1'),
(2,'ref_val_2'),
(3,'no_ref');
INSERT INTO zzz_dummy.subtable1 VALUES
(1,'1'),
(2,'2'),
(3,'3');
INSERT INTO zzz_dummy.maintable VALUES
(1,'1'),
(2,'2'),
(3,'1'),
(4,'1'),
(5,'2'),
(6,'1');
This will produce the following tables and entries:
maintable:
+----+-------------+
| id | ref_subtab1 |
+----+-------------+
| 1 | 1 |
| 3 | 1 |
| 4 | 1 |
| 6 | 1 |
| 2 | 2 |
| 5 | 2 |
+----+-------------+
subtable1:
+----+-------------+
| id | ref_subtab2 |
+----+-------------+
| 1 | 1 |
| 2 | 2 |
| 3 | 3 |
+----+-------------+
subtable2:
+----+-----------+
| id | col1 |
+----+-----------+
| 1 | ref_val_1 |
| 2 | ref_val_2 |
| 3 | no_ref |
+----+-----------+
PROBLEM
As you can see, the column ref_subtab1 in maintable references id in subtable1, which column ref_subtab2 finally references id in subtable2. I want to select all rows in subtable2 that are indirectly referenced in aforementioned manner.
I have tried
SELECT subtable2.* FROM zzz_dummy.subtable2
INNER JOIN zzz_dummy.maintable
INNER JOIN zzz_dummy.subtable1
WHERE zzz_dummy.maintable.ref_subtab1=zzz_dummy.subtable1.id
AND zzz_dummy.subtable1.ref_subtab2=zzz_dummy.subtable2.id;
but this returns 6 results, one for every match in maintable:
+----+-----------+
| id | col1 |
+----+-----------+
| 1 | ref_val_1 |
| 1 | ref_val_1 |
| 1 | ref_val_1 |
| 1 | ref_val_1 |
| 2 | ref_val_2 |
| 2 | ref_val_2 |
+----+-----------+
I do not want redundant values, I would like it to return:
+----+-----------+
| id | col1 |
+----+-----------+
| 1 | ref_val_1 |
| 2 | ref_val_2 |
+----+-----------+
Can this be done efficiently with a MySQL statement?

Answer 1

As already commented use distinct to get the unique result combination and also move those conditions from WHERE clause to JOIN ON condition like
SELECT distinct subtable2.*
FROM zzz_dummy.subtable2
INNER JOIN zzz_dummy.subtable1
ON zzz_dummy.subtable1.ref_subtab2 = zzz_dummy.subtable2.id
INNER JOIN zzz_dummy.maintable
ON zzz_dummy.maintable.ref_subtab1 = zzz_dummy.subtable1.id;