CSE 480: Database Systems

Lecture 11: SQL

1
SQL Query

SELECT <attribute list>

FROM <table list>
WHERE <condition that must be satisfied by a tuple>

– In MySQL, FROM and WHERE clauses are optional

– Example:

2
SELECT Clause

SELECT <attribute list>

 Attribute list may include

– * (wildcard)
– Keywords such as AS and DISTINCT
– Arithmetic expression
– String functions
– Boolean expression

3
Wildcard(*) in SELECT-Clause

 Retrieve the values for all columns of the selected tuples

4
AS in SELECT-Clause

 Rename the columns in query result

 Example: Find the names of employees who earn more
than their supervisors

SELECT E.FNAME AS EMP_FNAME, E.LNAME AS EMP_LNAME

FROM EMPLOYEE E, EMPLOYEE S
WHERE E.SUPERSSN=S.SSN AND E.SALARY >
S.SALARY;

Output: EMP_FNAME EMP_LNAME

5
DISTINCT in SELECT-Clause

 Eliminate duplicate tuples in query result

6
Arithmetic Expression in SELECT-Clause

 Query: Show the effect of giving employees a 10% raise

7
String functions in SELECT-Clause

8
String functions in SELECT-Clause

9
String functions in SELECT-Clause

For more string functions, go to

http://dev.mysql.com/doc/refman/4.1/en/string-functions.html

10
Boolean Expression in SELECT-Clause

11
FROM Clause

SELECT <attribute list>

FROM <table list>

 Table list may include

– Names of 1 or more tables
– Subquery for joined tables

12
Joined Relations in FROM-Clause
 Query 1: Retrieve the name and address of all employees who work
for the 'Research' department

SELECT Fname, Lname, Address

FROM EMPLOYEE, DEPARTMENT
WHERE Dname='Research' AND Dno=Dnumber

is equivalent to:

SELECT Fname, Lname, Address

FROM (EMPLOYEE JOIN DEPARTMENT ON Dno=Dnumber)
WHERE Dname='Research'

13
Joined Relations in FROM-Clause

 Many types of Join

– regular JOIN
– NATURAL JOIN
– CROSS JOIN
– LEFT OUTER JOIN
– RIGHT OUTER JOIN,
– etc

14
JOIN

R SELECT *
FROM R JOIN S on R.Id = S. Id S
Id Name
Id Value
111 John
foreach row r in table R 111 B
222 Mary
foreach row s in table S 111 A
333 Bill
if r.Id = s.Id then 222 A
444 Joe
output the merged
row of r and s

Id Name Id Value
111 John 111 B
111 John 111 A
222 Mary 222 A

15
JOIN – Example

Student Transcript
Id Name Addr Status StudId CrsCode Sem Grade
111 John ….. ….. 111 CSE305 S00 B
222 Mary ….. ….. 111 CSE306 S99 A
333 Bill ….. ….. 222 CSE304 F99 A
444 Joe ….. …..

SELECT * FROM Student JOIN Transcript on Id = StudId

Id Name Addr Status StudId CrsCode Sem Grade
111 John ….. ….. 111 CSE305 S00 B
111 John ….. ….. 111 CSE306 S99 A
222 Mary ….. ….. 222 CSE305 F99 A

Produces columns (attributes) with identical values, which is redundant

16
Natural Join

 Join condition equates all attributes with the same name

 Duplicate columns are automatically eliminated from result

R SELECT *
FROM R NATURAL JOIN S S
Id Name
foreach row r in table R Id Value
111 John
foreach row s in table S 111 B
222 Mary Examine their common attributes 111 A
333 Bill If their values are the same, then 222 A
444 Joe merge the rows while removing
their duplicate columns

Id Name Value
111 John B
111 John A
222 Mary A
17
Natural Join
Join attribute must have the same name (Id)
Student Transcript2
Id Name Addr Status Id CrsCode Sem Grade
111 John ….. ….. 111 CSE305 S00 B
222 Mary ….. ….. 111 CSE306 S99 A
333 Bill ….. ….. 222 CSE304 F99 A
444 Joe ….. …..

SELECT * FROM Student NATURAL JOIN Transcript2

Id Name Addr Status CrsCode Sem Grade
111 John ….. ….. CSE305 S00 B Duplicate attribute
111 John ….. ….. CSE306 S99 A (Id) was removed
222 Mary ….. ….. CSE305 F99 A

18
Cross Join

R SELECT *
FROM R CROSS JOIN S S
Id Name
Id Value
111 John
foreach row r in table R 111 B
222 Mary
foreach row s in table S 111 A
333 Bill
output the merged row r and s 222 A
444 Joe

Id Name Id Value
111 John 111 B
111 John 111 A
111 John 222 A
Output has 12 rows
… … … …
444 Joe 222 A
19
Left Outer Join
SELECT *
R FROM R LEFT OUTER JOIN S S
Id Name ON R.Id = S.Id
Id Value
111 John foreach row r in table R 111 B
222 Mary foreach row s in table S. 111 A
333 Bill if r.Id = S.Id then
222 A
444 Joe output the merged row r and s
if row r is not merged with any rows in S
output row r with NULL values for s

Id Name Id Value
111 John 111 B
111 John 111 A Output has 5 rows
222 Mary 222 A
333 Bill NULL NULL
444 Joe NULL NULL
20
Left Outer Join - Example

 Retrieve the name of all employees and their supervisors

SELECT E.Fname, E.Lname, S.Fname, S.Lname

FROM (EMPLOYEE E LEFT OUTER JOIN
EMPLOYEE S ON E.SUPERSSN=S.SSN)

21
Right Outer Join
SELECT *
R FROM R RIGHT OUTER JOIN S S
Id Name ON R.Id = S.Id
Id Value
111 John foreach row s in table S 111 B
222 Mary foreach row r in table R 111 A
333 Bill if r.Id = S.Id then
555 A
444 Joe output the merged row r and s
if row s is not merged with any rows in R
output row s with NULL values for r

Id Name Id Value
111 John 111 B
111 John 111 A Output has 3 rows
NULL NULL 555 A

22
WHERE-Clause

WHERE <condition>

 Selection condition is a Boolean expression

– Simple selection condition:
 <attribute> operator <constant>
 <attribute> operator <attribute>
 <attribute> operator <set> or <attribute> operator <relation>

– Complex conditions:
 <condition> AND <condition>
 <condition> OR <condition>
 NOT <condition>

23
Boolean Expression in WHERE-Clause

 <attribute> operator <constant>

 <attribute> operator <attribute>

– Operator: =, >, <, >=, <=, <> (not equal to)

– Applicable to integers, floats, strings, dates, etc. (except for
NULL)

SELECT *
FROM EMPLOYEE
WHERE SSN > SUPERSSN AND SALARY > 49999.99
AND MINIT='B' AND LNAME='SMITH'
AND BDATE >= '1980-01-01';

24
Substring Comparison in WHERE-Clause

 Find employees who live in “Houston, TX”.

 Use the LIKE operator to compare partial strings

 Two reserved characters are used:

– % matches an arbitrary number of characters
– _ matches a single arbitrary character
25
Substring Comparison

 Query: Retrieve all employees whose address is in

Houston, Texas.

SELECT FNAME, LNAME

FROM EMPLOYEE
WHERE ADDRESS LIKE '%Houston, TX%‘;

26
Substring Comparison

 Query: Retrieve all employees who were born during the

1950s.

Q26: SELECT FNAME, LNAME

FROM EMPLOYEE
WHERE BDATE LIKE '195_ - _ _ - _ _‘;

27
Arithmetic Expression in WHERE-Clause

 Query: Retrieve the names of employees who earn more

than half the salary of their supervisors

SELECT E.FName, E.Lname

FROM EMPLOYEE AS E, EMPLOYEE AS S
WHERE E.Salary > S.Salary/2
AND E.SuperSSN = S.SSN;

28
Arithmetic Expression in WHERE-Clause

 Between comparison operator

 Query: Retrieve the first and last names of employees in

department 5 whose salary is between $30,000 and
$40,000.

SELECT Fname, Lname

FROM EMPLOYEE
WHERE (Salary BETWEEN 30000 AND 40000)
AND Dno = 5

29
UNSPECIFIED WHERE-clause

 If there is only one relation in the FROM-clause and there

is no join condition, this implies all tuples of the relation
are selected

SELECT SSN
FROM EMPLOYEE;

 If more than one relation is specified in the FROM-clause

and there is no join condition, then the CARTESIAN
PRODUCT (Cross Join) of tuples is selected

SELECT SSN, DNAME

FROM EMPLOYEE, DEPARTMENT;

30
NULLS IN SQL QUERIES

 Cannot use equality (=) comparison to check for null values

 Query: Retrieve the names of employees who do not have

supervisors
SELECT FNAME, LNAME
FROM EMPLOYEE
WHERE SUPERSSN IS
NULL;
 Query: Retrieve the names of employees who have supervisors

SELECT FNAME, LNAME

FROM EMPLOYEE
WHERE SUPERSSN IS NOT
NULL;
31
SET OPERATIONS

 SQL has directly incorporated some set operations

– UNION
– INTERSECT
– MINUS or EXCEPT

 The resulting relations of these set operations are sets of tuples;

duplicate tuples are eliminated from the result

 The set operations are applicable to union compatible relations

– the two SQL relations must have the same attributes and the attributes
must appear in the same order

 Note: MySQL supports only UNION; Oracle supports all 3 set

operations

32
Example in Oracle

Duplicate rows are eliminated

Not available
in MySQL

33
Example

 Query: List the names of projects that involve an

employee whose last name is 'Smith' as a worker or as a
manager of the department that controls the project

34
Example

 Query: List the names of projects that involve an

employee whose last name is 'Smith' as a worker or as a
manager of the department that controls the project

(SELECT PNAME
FROM PROJECT, DEPARTMENT, EMPLOYEE
WHERE DNUM=DNUMBER AND MGRSSN=SSN
AND LNAME='Smith')
UNION
(SELECT PNAME
FROM PROJECT, WORKS_ON, EMPLOYEE
WHERE PNUMBER=PNO AND ESSN=SSN
AND NAME='Smith');

35
Example

 Query: List the first name and last name of employees

who do not work on any project

36
Example

 Query: List the first name and last name of employees

who do not work on any project

(SELECT Fname, Lname

FROM EMPLOYEE)
MINUS
(SELECT Fname, Lname
FROM WORKS_ON, EMPLOYEE
WHERE ESSN=SSN);

Caution: Not applicable in MySQL

(see slide 40 on how to write this query in MySQL)

37
IN Operator

v IN W

 The comparison operator IN compares a value v with a

set of values W, and evaluates to TRUE if v is one of the
elements in W. This is SET membership test.

 Examples:
– 3 in {1, 2, 3} TRUE
– 0 in {1, 2, 3} FALSE

38
IN Operator

 Query: Retrieve the social security numbers of all

employees who work on project number 1, 2, or 3

SELECT DISTINCT ESSN

FROM WORKS_ON
WHERE (PNO = 1) OR (PNO = 2) OR (PNO = 3);

Using IN Operator:

SELECT DISTINCT ESSN

FROM WORKS_ON
WHERE PNO IN (1, 2, 3);

39
Example

 Query: List the first name and last name of employees

who do not work on any project

SELECT Fname, Lname

FROM EMPLOYEE
WHERE SSN NOT IN
(SELECT ESSN FROM WORKS_ON);

40
Exercise

 Find the names of supervisors who are also managers

 Find the names of employees who are not supervisors
 Find the names of supervisors who supervise exactly one
employee

 Try to write it in two ways:

– Using IN operator
– Using SET operations (UNION, INTERSECT, MINUS)

41