Stuck with sql subquery - mysql

Question

I have these 3 tables :
1) Sailors (sid:INT, sname:VARCHAR(30), rating:INT, age:INT)
2) Boats (bid:INT, bname:VARCHAR(30), color:VARCHAR(10))
3) Reserves (bid:INT, sid:INT, day:DATE)
I want to write a query which displays the number(bid) of every boat that has been reserved by all the sailors at least once/
So far I wrote this :
SELECT b.bid, b.bname
FROM boats as b , sailors as s , reserves as r
WHERE r.bid = b.bid and r.sid = s.sid
GROUP BY r.bid
HAVING count(r.sid) = count(s.sid)
It seems the right answer to me but after trying it on MYSQL id didn't work.

Answer 1

Have you tried comparing the count of boat reserves to count of sailors like this?
HAVING count(r.sid) = (Select count(sid) from sailors)
you also need to make sure that count(r.sid) is a unique count of bid and sid incase the same sailor reserved the same boat more than once.
SELECT g.bid,
g.name
FROM
(SELECT b.bid, b.name, s.sid
FROM boats AS b, reserves AS r, sailors AS s
WHERE b.bid = r.bid AND r.sid = s.sid
GROUP BY b.bid, b.name, s.sid
) AS g
GROUP BY g.bid, g.name
HAVING COUNT(g.bid) = (SELECT COUNT(s.SID) FROM sailors AS s)

Answer 2

There are several query patterns that will return the specified result.
Assuming that the sid column in the Reserves table is a FOREIGN KEY referencing Sailors(sid), we can use a distinct count of sid from the Reserves table for a given boat, and compare that to a distinct count of sid from Sailors.
For example:
SELECT b.bid
, b.bname
FROM Boats b
JOIN ( SELECT rr.bid
, COUNT(DISTINCT rr.sid) AS cnt
FROM Reserves rr
GROUP BY rr.bid
) r
ON r.bid = b.bid
JOIN ( SELECT COUNT(DISTINCT ss.sid) AS cnt
FROM Sailors ss
) s
ON s.cnt = r.cnt
The inline view aliased as r gets us a count of distinct sid values for each boat, from the Reserves table.
The inline view aliased as s gets us a count of distinct sid values from Sailors. If sid is the PRIMARY KEY or UNIQUE KEY of Sailors (which is what we expect, it's the normative pattern) we can omit the DISTINCT keyword. The question did not indicate if this was the case, so rather than making a wrong assumption, we write a query that will work in the more general case.)
This query did assume that there was a foreign key constraint... that rows in Reserves would not have a non-NULL value that didn't reference a row in Sailors. If that's not the case, we can modify the inline view query r, so we get a distinct count of sid values that appear in the Sailors table.
This can replace the inline view query r above:
( SELECT rr.bid
, COUNT(DISTINCT rr.sid) AS cnt
FROM Reserves rr
JOIN Sailors rs
ON rs.sid = rr.sid
GROUP BY rr.bid
) r
This query pattern can easily be adapted to check conditions other than ALL sailors.
And there are other query patterns that can achieve an equivalent result.

Answer 3

You Can Check This out :)
===============================================
CREATE TABLE Sailors (sid INT, sname VARCHAR(30), rating INT, age INT)
INSERT INTO Sailors values (1,'S1',100,20);
INSERT INTO Sailors values (2,'S2',100,20);
INSERT INTO Sailors values (3,'S3',100,20);
INSERT INTO Sailors values (4,'S4',100,20);
INSERT INTO Sailors values (5,'S5',100,20);
CREATE TABLE Boats (bid INT, bname VARCHAR(30), color VARCHAR(10))
INSERT INTO Boats values (1,'B1','Blue');
INSERT INTO Boats values (2,'B2','Red');
INSERT INTO Boats values (3,'B3','Green');
INSERT INTO Boats values (4,'B4','Blue');
INSERT INTO Boats values (5,'B5','Green');
CREATE TABLE Reserves (bid INT, sid INT, day DATE)
INSERT INTO Reserves values (1,1,'2015-01-01');
INSERT INTO Reserves values (1,2,'2015-01-01');
INSERT INTO Reserves values (1,3,'2015-01-01');
INSERT INTO Reserves values (1,4,'2015-01-01');
INSERT INTO Reserves values (1,5,'2015-01-01');
INSERT INTO Reserves values (2,1,'2015-01-01');
INSERT INTO Reserves values (2,2,'2015-01-01');
INSERT INTO Reserves values (2,3,'2015-01-01');
INSERT INTO Reserves values (2,4,'2015-01-01');
INSERT INTO Reserves values (2,5,'2015-01-01');
INSERT INTO Reserves values (2,3,'2015-01-02');
INSERT INTO Reserves values (2,4,'2015-01-03');
INSERT INTO Reserves values (2,5,'2015-01-04');
INSERT INTO Reserves values (3,1,'2015-01-02');
INSERT INTO Reserves values (4,1,'2015-01-03');
INSERT INTO Reserves values (5,1,'2015-01-04');
===============================================
SELECT Reserves.bid , Boats.bname FROM Reserves , Boats
WHERE Reserves.bid = Boats.bid
GROUP BY Reserves.bid , Boats.bname
HAVING COUNT (DISTINCT Reserves.sid) = (SELECT COUNT(sid) FROM Sailors)
Result ===>
BID BNAME
---------------
1 B1
2 B2
Hope it will help :)