Multiple tags search query

Question

I working on a tag based search. I have three tables tag(id,name), tagXmedia(id,tag_id,media_id), and media(id,...). tagXmedia is the mapping table between the tag and media

Answer 1

SELECT
    media1.media_id
FROM
    (
        SELECT media_id FROM tagXmedia A INNER JOIN
        (SELECT id tag_id FROM tag WHERE name='home') B
        USING (tag_id)
    ) media1
    INNER JOIN
    (
        SELECT media_id FROM tagXmedia C INNER JOIN
        (SELECT id tag_id FROM tag WHERE name='hawaii') D
        USING (tag_id)
    ) media2
    USING (media_id)
;

Make sure you have this index in tagXmedia:

ALTER TABLE tagXmedia ADD UNIQUE INDEX (tag_id,media_id);

Here is a test case:

drop database if exists tagmediatest;
create database tagmediatest;
use tagmediatest
CREATE TABLE media
(
   id int not null auto_increment,
   stuff varchar(20),
   PRIMARY KEY (id)
);
INSERT INTO media (stuff) VALUES
('magazine'),('television'),('iphone'),
('ipad'),('IE9 Browser'),('radio');
CREATE TABLE tag
(
   id int not null auto_increment,
   name varchar(20),
   PRIMARY KEY (id),
   UNIQUE KEY (name)
);
INSERT INTO tag (name) VALUES
('away'),('home'),('jersery city'),('hawaii'),('nyc');
CREATE TABLE tagXmedia
(
   id int not null auto_increment,
   tag_id INT NOT NULL,
   media_id INT NOT NULL,
   PRIMARY KEY (id),
   UNIQUE KEY (tag_id,media_id)
);
INSERT INTO tagXmedia (tag_id,media_id) VALUES
(1,1),(1,2),(1,3),(1,6),
(2,1),(2,2),(2,4),(2,5),
(3,5),(3,4),(3,3),(3,1),
(4,2),(4,3),(4,5),(4,6),
(5,2),(5,3),(5,5),(5,4);
SELECT 
    media1.media_id 
FROM 
    ( 
        SELECT media_id FROM tagXmedia A INNER JOIN 
        (SELECT id tag_id FROM tag WHERE name='home') B 
        USING (tag_id) 
    ) media1 
    INNER JOIN 
    ( 
        SELECT media_id FROM tagXmedia C INNER JOIN 
        (SELECT id tag_id FROM tag WHERE name='hawaii') D 
        USING (tag_id) 
    ) media2
USING (media_id)
; 

Here is the result:

mysql> drop database if exists tagmediatest;
Query OK, 3 rows affected (0.09 sec)

mysql> create database tagmediatest;
Query OK, 1 row affected (0.00 sec)

mysql> use tagmediatest
Database changed
mysql> CREATE TABLE media
    -> (
    ->    id int not null auto_increment,
    ->    stuff varchar(20),
    ->    PRIMARY KEY (id)
    -> );
Query OK, 0 rows affected (0.05 sec)

mysql> INSERT INTO media (stuff) VALUES
    -> ('magazine'),('television'),('iphone'),
    -> ('ipad'),('IE9 Browser'),('radio');
Query OK, 6 rows affected (0.05 sec)
Records: 6  Duplicates: 0  Warnings: 0

mysql> CREATE TABLE tag
    -> (
    ->    id int not null auto_increment,
    ->    name varchar(20),
    ->    PRIMARY KEY (id),
    ->    UNIQUE KEY (name)
    -> );
Query OK, 0 rows affected (0.08 sec)

mysql> INSERT INTO tag (name) VALUES
    -> ('away'),('home'),('jersery city'),('hawaii'),('nyc');
Query OK, 5 rows affected (0.06 sec)
Records: 5  Duplicates: 0  Warnings: 0

mysql> CREATE TABLE tagXmedia
    -> (
    ->    id int not null auto_increment,
    ->    tag_id INT NOT NULL,
    ->    media_id INT NOT NULL,
    ->    PRIMARY KEY (id),
    ->    UNIQUE KEY (tag_id,media_id)
    -> );
Query OK, 0 rows affected (0.06 sec)

mysql> INSERT INTO tagXmedia (tag_id,media_id) VALUES
    -> (1,1),(1,2),(1,3),(1,6),
    -> (2,1),(2,2),(2,4),(2,5),
    -> (3,5),(3,4),(3,3),(3,1),
    -> (4,2),(4,3),(4,5),(4,6),
    -> (5,2),(5,3),(5,5),(5,4);
Query OK, 20 rows affected (0.05 sec)
Records: 20  Duplicates: 0  Warnings: 0

mysql> SELECT
    ->     media1.media_id
    -> FROM
    ->     (
    ->         SELECT media_id FROM tagXmedia A INNER JOIN
    ->         (SELECT id tag_id FROM tag WHERE name='home') B
    ->         USING (tag_id)
    ->     ) media1
    ->     INNER JOIN
    ->     (
    ->         SELECT media_id FROM tagXmedia C INNER JOIN
    ->         (SELECT id tag_id FROM tag WHERE name='hawaii') D
    ->         USING (tag_id)
    ->     ) media2
    -> USING (media_id)
    -> ;
+----------+
| media_id |
+----------+
|        2 |
|        5 |
+----------+
2 rows in set (0.00 sec)

mysql>

Note that tag_id 2 and 4 reside in media_id 2 and 5. This is why the query works.