Reputation: 1078
I wish to know which team in "TX" have not played a game. (In other words Im looking for a selection of records where there is no related record in the many table.)
Here is the SQL:
(Or if You prefer the sql fiddle is here:http://sqlfiddle.com/#!2/14106 )
CREATE TABLE `Team` (
`ID` INT NOT NULL AUTO_INCREMENT ,
`Name` VARCHAR(45) NULL ,
`State` VARCHAR(45) NULL ,
PRIMARY KEY (`ID`) );
CREATE TABLE `Games` (
`ID` INT NOT NULL AUTO_INCREMENT,
`Team_ID` INT NULL ,
`Game_Day` DATE NULL ,
PRIMARY KEY (`ID`) );
INSERT INTO `Team` (`Name`, `State`) VALUES ('Rams', 'TX');
INSERT INTO `Team` (`Name`, `State`) VALUES ('Rockets', 'OK');
INSERT INTO `Team` (`Name`, `State`) VALUES ('Bombers', 'TX');
INSERT INTO `Team` (`Name`, `State`) VALUES ('Yellow Jackets', 'NV');
INSERT INTO `Team` (`Name`, `State`) VALUES ('Wildcats', 'CT');
INSERT INTO `Team` (`Name`, `State`) VALUES ('Miners', 'CO');
INSERT INTO `Team` (`Name`, `State`) VALUES ('Bolts', 'TX');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('2', '2013-03-16');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('2', '2013-01-01');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('3', '2013-04-16');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('5', '2013-02-02');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('4', '2013-02-12');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('6', '2013-01-09');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('6', '2013-01-01');
INSERT INTO `Games` (`Team_ID`, `Game_Day`) VALUES ('3', '2013-05-01');
I should get the result:
ID Name
1 Rams
7 Bolts
Upvotes: 5
Views: 2409
Reputation: 6461
SELECT `ID`, `Name` FROM `TEAM`
WHERE `ID` NOT IN (SELECT DISTINCT(`Team_ID`) from `Games`)
AND `State` = 'TX';
SqlFiddle here.
Upvotes: 6
Reputation: 425378
Use an outer join, selecting only those rows that don't match
SELECT t.*
FROM TEAM t
LEFT JOIN GAMES g ON g.team_id = t.id
WHERE t.state = 'TX'
AND g.team_id is null -- return only rows that *don't* join
This this running in SQL Fiddle
Note that using a join will out-perform a sub-query approach, especially when data sets become large.
Upvotes: 6
Reputation: 26363
You can also left-join to the Games
table and filter for where there isn't a corresponding Games
row. This is usually faster than NOT IN
when the tables have a lot of rows:
SELECT Team.ID, Team.Name
FROM Team
LEFT JOIN Games ON Team.ID = Games.Team_ID
WHERE Team.State = 'TX' AND Games.ID IS NULL;
If there isn't a Games
row to go with the Teams
row, the Games.ID
column will be null in the result, so if you filter on Games.ID IS NULL
you'll get all the rows where a team has no games.
There's a SQL Fiddle here.
Upvotes: 4
Reputation: 310
Hope this would help.
select t.ID,t.NAME
FROM Team t
WHERE t.state = 'TX'
AND t.id NOT IN (SELECT g.TEAM_ID FROM Games g)
Upvotes: 1