Reputation: 53
I am trying to find student who has min score which will be the result of the below query. However, I was asked to write the query without using MIN(). Spent several hours but I can't find the alternative solution :'(.
select s.sname
from student s
where s.score =
(select min(s2.score)
from score s2)
Upvotes: 0
Views: 6721
Reputation: 310
One way of doing it would be to Order the results in Ascending order and take the first row. But if you are looking at a more generic solution as a student will have more than one mark associated with him, So you need to find the total marks for each student and then find the student with the least total.
This is the first scenario, A student only has one row in the table.
CREATE TABLE Student
(
SLNO INT,
MARKS FLOAT,
NAME NVARCHAR(MAX)
)
INSERT INTO Student VALUES(1, 80, 't1')
INSERT INTO Student VALUES(2, 90, 't2')
INSERT INTO Student VALUES(3, 76, 't3')
INSERT INTO Student VALUES(4, 98, 't4')
INSERT INTO Student VALUES(5, 55, 't5')
SELECT * From Student ORDER BY MARKS ASC
The second scenario as specified above is, He has multiple rows in the table, So we insert two more rows into the table for existing users.
Then we select the users by taking the sum of their marks grouping the results by name and then ordering the results by their total
INSERT INTO Student VALUES(6, 55, 't1')
INSERT INTO Student VALUES(6, 90, 't5')
SELECT SUM(MARKS) AS TOTAL, NAME FROM Student
GROUP BY NAME
ORDER BY TOTAL
Hope the above is what you are looking for.
Upvotes: 1
Reputation: 12164
create table student (name varchar(10), score int);
insert into student (name, score) values('joe', 30);
insert into student (name, score) values('jim', 88);
insert into student (name, score) values('jack', 22);
insert into student (name, score) values('jimbo', 15);
insert into student (name, score) values('jo bob',15);
/* folks with lowest score */
select name, score from student where not exists(select 1 from student s where s.score < student.score);
/* the actual lowest score */
select distinct score from student
where not exists(select 1 from student s where s.score < student.score);
Note that not exists can be brutally inefficient, but it'll do the job on a small set.
Upvotes: 1
Reputation:
Here's a possible alternative to the JOIN
approach:
select sname from student where score in
(select score from student order by score asc limit 1)
Upvotes: 1
Reputation: 13425
This is one way, which will work even if two students have same lowest score.
SELECT distinct s1.sname
FROM student s1
LEFT JOIN student s2
ON s2.score < s1.score
WHERE s2.score IS NULL
The below is the method using limit, which will return lowest score student, but only one of them if multiple of them have same score.
select sname
from student
order by score asc
limit 1
Upvotes: 3