SQL Queries and Subqueries

Basic SQL Query

SELECT [DISTINCT] target-list

FROM relation-list
WHERE qualification;

• Relation-list: A list of relation names (possibly with range-

variable after each name).
• Target-list: A list of attributes of relations in relation-list
• Qualification: conditions on attributes
• DISTINCT: optional keyword for duplicate removal.
– Default = no duplicate removal!
SQL Comparison Operators
 How to evaluate a query?

SELECT [DISTINCT] target-list

FROM relation-list
WHERE qualification;

• Conceptual query evaluation using relational operators:

1) Compute the cross-product of relation-list.
2) Discard resulting tuples if they fail qualifications.
3) Delete attributes that are not in target-list. (called column-list)
4) If DISTINCT is specified, eliminate duplicate rows.

SELECT S.sname
FROM Sailors S, Reserves R
WHERE S.sid=R.sid AND R.bid=103;
 Example of Conceptual Evaluation (1)

SELECT S.sname (1) Compute the cross-product of relation-

FROM Sailors S, Reserves R list.
WHERE S.sid=R.sid AND R.bid=103;

Sailors Reserves

sid sname rating age sid bid day

22 dustin 7 45.0 22 101 10/10/96
X
31 lubber 8 55.5 58 103 11/12/96
58 rusty 10 35.0
 Example of Conceptual Evaluation (2)
SELECT S.sname (2) Discard tuples if they fail
FROM Sailors S, Reserves R qualifications.
WHERE S.sid=R.sid AND R.bid=103;
Sailors X Reserves
S.sid sname rating age R.sid bid day
22 dustin 7 45.0 22 101 10/10/96
22 dustin 7 45.0 58 103 11/12/96
31 lubber 8 55.5 22 101 10/10/96
31 lubber 8 55.5 58 103 11/12/96
58 rusty 10 35.0 22 101 10/10/96
58 rusty 10 35.0 58 103 11/12/96
 Example of Conceptual Evaluation (3)
SELECT S.sname (3) Delete attribute columns that are
FROM Sailors S, Reserves R not in target-list.
WHERE S.sid=R.sid AND R.bid=103;
sname
Sailors X Reserves rusty

(sid) sname rating age (sid) bid day

22 dustin 7 45.0 22 101 10/10/96
22 dustin 7 45.0 58 103 11/12/96
31 lubber 8 55.5 22 101 10/10/96
31 lubber 8 55.5 58 103 11/12/96
58 rusty 10 35.0 22 101 10/10/96
58 rusty 10 35.0 58 103 11/12/96
 Renaming / Aliasing

Consider the following SALESREPS relation

Empl_num name age Rep_office manager

105 Bill 37 13 104
104 Bob 33 12 106
106 Sam 52 11 NULL

How do we determine the name of Bob’s manager?

 Aliasing

SELECT s2.name
FROM SALESREPS s1, SALESREPS s2
WHERE s1.name=‘Bob’ AND
s1.manager=s2.empl_num;

• Aliases must be used here.

• The row referenced by s1 is intended to be Bob…
• …while s2 will be his manager’s.
• Remember, first FROM, then WHERE, then SELECT
 Relational Design Example
• Students (PID: string, Name: string, Address: string)

• Professors (PID: string, Name: string, Office: string, Age: integer, DepartmentName:
string)

• Courses (Number: integer, DeptName: string, CourseName: string, Classroom:

string, Enrollment: integer)

• Teach (ProfessorPID: string, Number: integer, DeptName: string)

• Take (StudentPID: string, Number: integer, DeptName: string, Grade: string,

ProfessorEvaluation: integer)

• Departments (Name: string, ChairmanPID: string)

• PreReq (Number: integer, DeptName: string, PreReqNumber: integer,

PreReqDeptName: string)
 Motivation for Subqueries
• Find the name of the professor who teaches “CS 4604.”
SELECT Name
FROM Professors, Teach
WHERE (PID = ProfessorPID) AND (Number = ‘4604’)
AND(DeptName = ‘CS’) ;

• Do we need to take the natural join of two big relations

just to get a relation with one tuple?
• Can we rewrite the query without using a join?
 Nesting

• A query can be put inside another query

• Most commonly in the WHERE clause
• Sometimes in the FROM clause (depending on the software)
• This subquery is executed first (if possible)
 Subquery Example

• Find the name of the professor who teaches “CS 4604.”

SELECT Name
FROM Professors
WHERE PID =
(SELECT ProfessorPID
FROM Teach
WHERE (Number = 4604) AND (DeptName = ’CS’)
);

• When using =, the subquery must return a single tuple

 Conditions Involving Relations
• SQL includes a number of operators that apply to a
relation and produce a boolean result.
• These operators are very useful to apply on results of sub-
queries.
• Let R be a relation and t be a tuple with the same set of
attributes.
– EXISTS R is true if and only if R contains at least one tuple.
– t IN R is true if and only if t equals a tuple in R.
– t > ALL R is true if and only if R is unary (has one attribute) and
t is greater than every value in R.
• Can use any of the other five comparison operators.
• If we use <>, R need not be unary.
– t > ANY R (which is unary) is true if and only if t is greater than at
least one value in R.
• We can use NOT to negate EXISTS, ALL, and ANY.
 Subqueries Using Conditions

• Find the departments of the courses taken by the student

with name ‘Suri’.

SELECT DeptName
FROM Take
WHERE StudentPID IN
( SELECT PID
FROM Students
WHERE (Name = ‘Suri’)
);
 Correlated vs Uncorrelated

• The previous subqueries did not depend on anything outside

the subquery
– …and thus need to be executed just once.
– These are called uncorrelated.

• A correlated subquery depends on data from the outer query

– … and thus has to be executed for each row of the outer table(s)
 Correlated Subqueries

• Find course names that have been used for two or more courses.

SELECT CourseName
FROM Courses AS First
WHERE CourseName IN
(SELECT CourseName
FROM Courses
WHERE (Number <> First.Number)
AND (DeptName <> First.DeptName)
);
Evaluating Correlated Subqueries
SELECT CourseName
FROM Courses AS First
WHERE CourseName IN
(SELECT CourseName
FROM Courses
WHERE (Number <> First.Number)
AND (DeptName <> First.DeptName)
);
• Evaluate query by looping over tuples of First, and for each tuple
evaluate the subquery.
• Scoping rules: an attribute in a subquery belongs to one of the
tuple variables in that subquery’s FROM clause, or to the
immediately surrounding subquery, and so on.
 Subqueries in FROM clauses
• Can use a subquery as a relation in a FROM clause.
• We must give such a relation an alias using the AS keyword.
• Let us find different ways of writing the query “Find the names of
Professors who have taught the student whose first name is
’Suri’.”
• The old way:
SELECT Professors.Name
FROM Professors, Take, Teach, Students
WHERE (Professors.PID = Teach.ProfessorPID)
AND (Teach.CourseNumber = Take.CourseNumber)
AND (Teach.DeptName = Take.DeptName)
AND (Take.StudentPID = Student.PID)
AND (Student.Name = ’Suri %’);
• “Find the names of (Professors who have taught (courses taken
by (student with first name ’Suri’))).”

SELECT Name
FROM Professors
WHERE PID IN
(SELECT ProfessorPID
FROM Teach
WHERE (Number, DeptName) IN
( SELECT Number, DeptName
FROM TakeWHERE StudentPID IN
(SELECT PID
FROM Students
WHERE Name = ’Suri %’)
)
);
 Aggregate Operators
• COUNT (*)
• COUNT ( [DISTINCT] A)
– A is a column
• SUM ( [DISTINCT] A)
• AVG ( [DISTINCT] A)
• MAX (A)
• MIN (A)
• Count the number of sailors
SELECT COUNT (*)
FROM Sailors S

21
Find the average age of sailors with
rating = 10
Sailors(sid: integer, sname: string, rating: integer,
age: real)

SELECT AVG (S.age)

FROM Sailors S
WHERE S.rating=10

22
 Count the number of different
sailor names
Sailors(sid: integer, sname: string, rating: integer,
age: real)
SELECT COUNT (DISTINCT S.sname)
FROM Sailors S

23
Find the age of the oldest sailor
Sailors(sid: integer, sname: string, rating: integer,
age: real)
SELECT MAX(S.AGE)
FROM Sailors S

24
 Find name and age of the oldest
sailor(s)
SELECT S.sname, MAX (S.age)
FROM Sailors S

• This is illegal, but why?

– Cannot combine a column with a value

SELECT S.sname, S.age

FROM Sailors S
WHERE S.age = (SELECT MAX (S2.age) FROM Sailors S2)

25
 GROUP BY and HAVING
• So far, aggregate operators are applied to all (qualifying)
tuples.
– Can we apply them to each of several groups of tuples?
• Example: find the age of the youngest sailor for each rating
level.
– In general, we don’t know how many rating levels exist, and what the
rating values for these levels are!
– Suppose we know that rating values go from 1 to 10; we can write 10
queries that look like this:

SELECT MIN (S.age)

For i = 1, 2, ... , 10: FROM Sailors S
WHERE S.rating = i
26
 Find the age of the youngest sailor
for each rating level
Sid Sname Rating Age
SELECT S.rating, MIN (S.age) as age
FROM Sailors S 22 Dustin 7 45.0
GROUP BY S.rating 31 Lubber 8 55.5
(1) The sailors tuples are put into “same 85 Art 3 25.5
rating” groups. 32 Andy 8 25.5
(2) Compute the Minimum age for each 95 Bob 3 63.5
rating group.
Rating Age
3 25.5 (1)
Rating Age 3 63.5
3 25.5 (2) 7 45.0
7 45.0 8 55.5
8 25.5 8 25.5 27
 Find the age of the youngest sailor for
each rating level that has at least 2
members Sid Sname Rating Age
SELECT S.rating, MIN (S.age) as 22 Dustin 7 45.0
minage
FROM Sailors S 31 Lubber 8 55.5
GROUP BY S.rating 85 Art 3 25.5
HAVING COUNT(*) > 1 32 Andy 8 25.5
1. The sailors tuples are put into 95 Bob 3 63.5
“same rating” groups.
2. Eliminate groups that have < 2 Rating Age
members. 3 25.5
3. Compute the Minimum age for 3 63.5
each rating group. Rating Minage 7 45.0
3 25.5 8 55.5
8 25.5 8 25.5
Queries With GROUP BY and HAVING
SELECT [DISTINCT] target-list
FROM relation-list
SELECT S.rating, MIN (S.age) as age
WHERE qualification
FROM Sailors S
GROUP BY grouping-list
GROUP BY S.rating
HAVING group-qualification
HAVING S.rating > 5

• The target-list contains (i) attribute names (ii) terms with

aggregate operations (e.g., AVG (S.age)).
• The attribute list (e.g., S.rating) in target-list must be in
grouping-list.
• The attributes in group-qualification must be in grouping-list.

29
 Starwars Exercises
char(name, race, homeworld, affiliation)
planets(name, type, affiliation)
timetable(cname, pname, movie, arrival, departure)

• Which planet does Princess Leia go to in movie3?

SELECT distinct pname

FROM timetable
WHERE cname ='Princess Leia' and movie=3;

30
 Starwars Exercises
char(name, race, homeworld, affiliation)
planets(name, type, affiliation)
timetable(cname, pname, movie, arrival, departure)

• How many humans stay on Dagobah in movie 3?

SELECT count(*)
FROM timetable, characters
WHERE movie=3 and pname =‘Dagobah’ and
timetable.cname=characters.name and
characters.race=‘Human’;

31
 Starwars Exercises
char(name, race, homeworld, affiliation)
planets(name, type, affiliation)
timetable(cname, pname, movie, arrival, departure)

• Who has been to his/her homeworld in movie 2?

SELECT distinct c.name

FROM characters c, timetable t
WHERE c.name=t.cname and t.pname=c.homeworld and
movie=2;

32
 Starwars Exercises
char(name, race, homeworld, affiliation)
planets(name, type, affiliation)
timetable(cname, pname, movie, arrival, departure)

• Find distinct names of the planets visited by those of race

“droid”.

SELECT distinct t.pname

FROM char c, timetable t
WHERE c.name=t.cname and c.race=‘droid’;

33
 Starwars Exercises
char(name, race, homeworld, affiliation)
planets(name, type, affiliation)
timetable(cname, pname, movie, arrival, departure)

• For each character and for each neutral planet, how much
time total did the character spend on the planet?

SELECT c.name, p.name, SUM(t.departure-t.arrival) as amount

FROM characters c, timetable t, planets p
WHERE t.cname=c.name and t.pname=p.name and
p.affiliation='neutral'
GROUP BY c.name, p.name;

34