CHAPTER 7

More SQL: Complex Queries,

Triggers, Views, and Schema
Modification

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 1

Chapter 7 Outline
 More Complex SQL Retrieval Queries
 Specifying Semantic Constraints as Assertions
and Actions as Triggers
 Views (Virtual Tables) in SQL
 Schema Modification in SQL

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 3

More Complex SQL Retrieval
Queries
 Additional features allow users to specify more
complex retrievals from database:
 Nested queries, joined tables, and outer joins (in
the FROM clause), aggregate functions, and
grouping

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 4

Comparisons Involving NULL
and Three-Valued Logic
 Meanings of NULL
 Unknown value
 Unavailable or withheld value
 Not applicable attribute
 Each individual NULL value considered to be
different from every other NULL value
 SQL uses a three-valued logic:
 TRUE, FALSE, and UNKNOWN (like Maybe)
 NULL = NULL comparison is avoided

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 5

Comparisons Involving NULL
and Three-Valued Logic (cont’d.)

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 6

Comparisons Involving NULL
and Three-Valued Logic (cont’d.)
 SQL allows queries that check whether an
attribute value is NULL
 IS or IS NOT NULL

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 7

Nested Queries, Tuples,
and Set/Multiset Comparisons
 Nested queries
 Complete select-from-where blocks within WHERE
clause of another query
 Outer query and nested subqueries
 Comparison operator IN
 Compares value v with a set (or multiset) of values
V
 Evaluates to TRUE if v is one of the elements in V

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 8

Nested Queries (cont’d.)

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 9

Nested Queries (cont’d.)
 Use tuples of values in comparisons
 Place them within parentheses

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 10

Nested Queries (cont’d.)
 Use other comparison operators to compare a
single value v
 = ANY (or = SOME) operator

Returns TRUE if the value v is equal to some value in
the set V and is hence equivalent to IN
 Other operators that can be combined with ANY (or
SOME): >, >=, <, <=, and <>
 ALL: value must exceed all values from nested
query

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 11

Nested Queries (cont’d.)
 Avoid potential errors and ambiguities
 Create tuple variables (aliases) for all tables
referenced in SQL query

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 11

Correlated Nested Queries
 Queries that are nested using the = or IN
comparison operator can be collapsed into one
single block: E.g., Q16 can be written as:

 Q16A: SELECT E.Fname, E.Lname

FROM EMPLOYEE AS E, DEPENDENT AS D
WHERE E.Ssn=D.Essn AND E.Sex=D.Sex
AND
E.Fname=D.Dependent_name;
 Correlated nested query
 Evaluated once for each tuple in the outer query

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 13

The EXISTS and UNIQUE Functions
in SQL for correlating queries
 EXISTS function
 Check whether the result of a correlated nested
query is empty or not. They are Boolean functions
that return a TRUE or FALSE result.
 EXISTS and NOT EXISTS
 Typically used in conjunction with a correlated
nested query
 SQL function UNIQUE(Q)
 Returns TRUE if there are no duplicate tuples in
the result of query Q

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 14

USE of EXISTS

Q7:

SELECT Fname, Lname

FROM Employee
WHERE EXISTS (SELECT *
FROM DEPENDENT
WHERE Ssn= Essn)

AND EXISTS (SELECT *

FROM Department
WHERE Ssn= Mgr_Ssn)

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 15

USE OF NOT EXISTS
To achieve the “for all” (universal quantifier- see Ch.8) effect,
we use double negation this way in SQL:
Query: List first and last name of employees who work on
ALL projects controlled by Dno=5.
SELECT Fname, Lname
FROM Employee
WHERE NOT EXISTS ( (SELECT Pnumber
FROM PROJECT
WHERE Dno=5)

EXCEPT (SELECT Pno

FROM WORKS_ON
WHERE Ssn= ESsn)
The above is equivalent to double negation: List names of those
employees for whom there does NOT exist a project managed by
department no. 5 that they do NOT work on.
Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 16
Double Negation to accomplish “for
all” in SQL
 Q3B: SELECT Lname, Fname
FROM EMPLOYEE
WHERE NOT EXISTS ( SELECT *
FROM WORKS_ON B
WHERE ( B.Pno IN ( SELECT Pnumber
FROM PROJECT
WHERE Dnum=5
AND

NOT EXISTS (SELECT *

FROM WORKS_ON C
WHERE C.Essn=Ssn
AND C.Pno=B.Pno )));
The above is a direct rendering of: List names of those employees for whom
there does NOT exist a project managed by department no. 5 that they
do NOT work on.

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 16

Explicit Sets and Renaming of
Attributes in SQL
 Can use explicit set of values in WHERE clause
Q17: SELECT DISTINCT Essn
FROM WORKS_ON
WHERE Pno IN (1, 2, 3);
 Use qualifier AS followed by desired new name
 Rename any attribute that appears in the result of
a query

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 18

Specifying Joined Tables in the
FROM Clause of SQL
 Joined table
 Permits users to specify a table resulting from a
join operation in the FROM clause of a query
 The FROM clause in Q1A
 Contains a single joined table. JOIN may also be
called INNER JOIN

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 19

Different Types of JOINed Tables in
SQL
 Specify different types of join
 NATURAL JOIN
 Various types of OUTER JOIN (LEFT, RIGHT,
FULL )
 NATURAL JOIN on two relations R and S
 No join condition specified
 Is equivalent to an implicit EQUIJOIN condition for
each pair of attributes with same name from R and
S

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 20

NATURAL JOIN
 Rename attributes of one relation so it can be joined with
another using NATURAL JOIN:

Q1B: SELECT Fname, Lname, Address

FROM (EMPLOYEE NATURAL JOIN
(DEPARTMENT AS DEPT (Dname, Dno, Mssn,
Msdate)))
WHERE Dname=‘Research’;

The above works with EMPLOYEE.Dno = DEPT.Dno as an

implicit join condition

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 20

INNER and OUTER Joins
 INNER JOIN (versus OUTER JOIN)
 Default type of join in a joined table

 Tuple is included in the result only if a matching tuple exists

in the other relation

 LEFT OUTER JOIN
 Every tuple in left table must appear in result

 If no matching tuple


Padded with NULL values for attributes of right table
 RIGHT OUTER JOIN
 Every tuple in right table must appear in result

 If no matching tuple


Padded with NULL values for attributes of left table

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 22

Example: LEFT OUTER JOIN

SELECT E.Lname AS Employee_Name

S.Lname AS Supervisor_Name

FROM Employee AS E LEFT OUTER JOIN EMPLOYEE AS S

ON E.Super_ssn = S.Ssn)

ALTERNATE SYNTAX:
SELECT E.Lname , S.Lname
FROM EMPLOYEE E, EMPLOYEE S
WHERE E.Super_ssn + = S.Ssn

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 22

Multiway JOIN in the FROM clause
 FULL OUTER JOIN – combines result if LEFT
and RIGHT OUTER JOIN
 Can nest JOIN specifications for a multiway join:

Q2A: SELECT Pnumber, Dnum, Lname, Address, Bdate

FROM ((PROJECT JOIN DEPARTMENT ON
Dnum=Dnumber) JOIN EMPLOYEE ON
Mgr_ssn=Ssn)
WHERE Plocation=‘Stafford’;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 23

Aggregate Functions in SQL
 Used to summarize information from multiple
tuples into a single-tuple summary
 Built-in aggregate functions
 COUNT, SUM, MAX, MIN, and AVG
 Grouping
 Create subgroups of tuples before summarizing
 To select entire groups, HAVING clause is used
 Aggregate functions can be used in the SELECT
clause or in a HAVING clause

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 24

Renaming Results of Aggregation
 Following query returns a single row of computed values
from EMPLOYEE table:

Q19: SELECT SUM (Salary), MAX (Salary), MIN (Salary), AVG

(Salary)
FROM EMPLOYEE;
 The result can be presented with new names:

Q19A: SELECT SUM (Salary) AS Total_Sal, MAX (Salary) AS

Highest_Sal, MIN (Salary) AS Lowest_Sal, AVG
(Salary) AS Average_Sal
FROM EMPLOYEE;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 25

Aggregate Functions in SQL (cont’d.)
 NULL values are discarded when aggregate
functions are applied to a particular column

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 26

Aggregate Functions on Booleans
 SOME and ALL may be applied as functions on
Boolean Values.
 SOME returns true if at least one element in the
collection is TRUE (similar to OR)
 ALL returns true if all of the elements in the
collection are TRUE (similar to AND)

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 27

Grouping: The GROUP BY Clause
 Partition relation into subsets of tuples
 Based on grouping attribute(s)
 Apply function to each such group independently
 GROUP BY clause
 Specifies grouping attributes
 COUNT (*) counts the number of rows in the
group

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 28

Examples of GROUP BY
 The grouping attribute must appear in the SELECT
clause:
Q24: SELECT Dno, COUNT (*), AVG (Salary)
FROM EMPLOYEE
GROUP BY Dno;
 If the grouping attribute has NULL as a possible value,
then a separate group is created for the null value (e.g.,
null Dno in the above query)
 GROUP BY may be applied to the result of a JOIN:
Q25: SELECT Pnumber, Pname, COUNT (*)
FROM PROJECT, WORKS_ON
WHERE Pnumber=Pno
GROUP BY Pnumber, Pname;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 29

Grouping: The GROUP BY and
HAVING Clauses (cont’d.)
 HAVING clause
 Provides a condition to select or reject an entire
group:
 Query 26. For each project on which more than two employees work,
retrieve the project number, the project name, and the number of
employees who work on the project.

Q26: SELECT Pnumber, Pname, COUNT (*)

FROM PROJECT, WORKS_ON
WHERE Pnumber=Pno
GROUP BY Pnumber, Pname
HAVING COUNT (*) > 2;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 31

Combining the WHERE and the
HAVING Clause
 Consider the query: we want to count the total number of
employees whose salaries exceed $40,000 in each
department, but only for departments where more than
five employees work.

 INCORRECT QUERY:
SELECT Dno, COUNT (*)
FROM EMPLOYEE
WHERE Salary>40000
GROUP BY Dno
HAVING COUNT (*) > 5;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 31

Combining the WHERE and the
HAVING Clause (continued)
Correct Specification of the Query:
 Note: the WHERE clause applies tuple by tuple

whereas HAVING applies to entire group of tuples

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 32

Use of WITH
 The WITH clause allows a user to define a table
that will only be used in a particular query (not
available in all SQL implementations)
 Used for convenience to create a temporary
“View” and use that immediately in a query
 Allows a more straightforward way of looking a
step-by-step query

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 33

Example of WITH
 See an alternate approach to doing Q28:
 Q28’: WITH BIGDEPTS (Dno) AS
( SELECT Dno
FROM EMPLOYEE
GROUP BY Dno
HAVING COUNT (*) > 5)
SELECT Dno, COUNT (*)
FROM EMPLOYEE
WHERE Salary>40000 AND Dno IN BIGDEPTS
GROUP BY Dno;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 34

Use of CASE
 SQL also has a CASE construct
 Used when a value can be different based on
certain conditions.
 Can be used in any part of an SQL query where a
value is expected
 Applicable when querying, inserting or updating
tuples

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 35

EXAMPLE of use of CASE
 The following example shows that employees are
receiving different raises in different departments
(A variation of the update U6)

 U6’: UPDATE EMPLOYEE

SET Salary =
CASE WHEN Dno = 5THEN Salary + 2000
WHEN Dno = 4THEN Salary + 1500
WHEN Dno = 1THEN Salary + 3000

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 36

Recursive Queries in SQL
 An example of a recursive relationship between
tuples of the same type is the relationship
between an employee and a supervisor.
 This relationship is described by the foreign key
Super_ssn of the EMPLOYEE relation
 An example of a recursive operation is to retrieve all supervisees of
a supervisory employee e at all levels—that is, all employees e
directly supervised by e, all employees e’ directly supervised by each
employee e, all employees e directly supervised by each employee
e, and so on. Thus the CEO would have each employee in the
company as a supervisee in the resulting table. Example shows such
table SUP_EMP with 2 columns (Supervisor,Supervisee(any level)):

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 37

An EXAMPLE of RECURSIVE Query
 Q29: WITH RECURSIVE SUP_EMP (SupSsn, EmpSsn) AS
SELECT SupervisorSsn, Ssn
FROM EMPLOYEE
UNION
SELECT E.Ssn, S.SupSsn
FROM EMPLOYEE AS E, SUP_EMP AS S
WHERE E.SupervisorSsn = S.EmpSsn)
SELECT *
FROM SUP_EMP;
 The above query starts with an empty SUP_EMP and
successively builds SUP_EMP table by computing
immediate supervisees first, then second level
supervisees, etc. until a fixed point is reached and no
more supervisees can be added
Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 38
EXPANDED Block Structure of SQL
Queries

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 40

Specifying Constraints as Assertions
and Actions as Triggers
 Semantic Constraints: The following are beyond
the scope of the EER and relational model
 CREATE ASSERTION
 Specify additional types of constraints outside
scope of built-in relational model constraints
 CREATE TRIGGER
 Specify automatic actions that database system
will perform when certain events and conditions
occur

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 41

Specifying General Constraints as
Assertions in SQL
 CREATE ASSERTION
 Specify a query that selects any tuples that violate
the desired condition
 Use only in cases where it goes beyond a simple
CHECK which applies to individual attributes and
domains

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 41

Introduction to Triggers in SQL
 CREATE TRIGGER statement
 Used to monitor the database
 Typical trigger has three components which make
it a rule for an “active database “ (more on active
databases in section 26.1) :
 Event(s)
 Condition
 Action

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 43

USE OF TRIGGERS
 AN EXAMPLE with standard Syntax.(Note : other
SQL implementations like PostgreSQL use a
different syntax.)

R5:
CREATE TRIGGER SALARY_VIOLATION
BEFORE INSERT OR UPDATE OF Salary, Supervisor_ssn ON
EMPLOYEE

FOR EACH ROW

WHEN (NEW.SALARY > ( SELECT Salary FROM EMPLOYEE
WHERE Ssn = NEW. Supervisor_Ssn))
INFORM_SUPERVISOR (NEW.Supervisor.Ssn, New.Ssn)

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 44

Views (Virtual Tables) in SQL
 Concept of a view in SQL
 Single table derived from other tables called the
defining tables
 Considered to be a virtual table that is not
necessarily populated

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 45

Specification of Views in SQL
 CREATE VIEW command
 Give table name, list of attribute names, and a query to
specify the contents of the view
 In V1, attributes retain the names from base tables. In
V2, attributes are assigned names

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 46

Specification of Views in SQL
(cont’d.)
 Once a View is defined, SQL queries can use the
View relation in the FROM clause
 View is always up-to-date
 Responsibility of the DBMS and not the user
 DROP VIEW command
 Dispose of a view

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 47

View Implementation, View Update,
and Inline Views
 Complex problem of efficiently implementing a
view for querying
 Strategy1: Query modification approach
 Compute the view as and when needed. Do not
store permanently
 Modify view query into a query on underlying base
tables
 Disadvantage: inefficient for views defined via
complex queries that are time-consuming to
execute

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 48

View Materialization
 Strategy 2: View materialization
 Physically create a temporary view table when the
view is first queried
 Keep that table on the assumption that other
queries on the view will follow
 Requires efficient strategy for automatically updating
the view table when the base tables are updated
 Incremental update strategy for materialized
views
 DBMS determines what new tuples must be inserted,
deleted, or modified in a materialized view table
Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 49
View Materialization (contd.)
 Multiple ways to handle materialization:
 immediate update strategy updates a view as
soon as the base tables are changed
 lazy update strategy updates the view when
needed by a view query
 periodic update strategy updates the view
periodically (in the latter strategy, a view query
may get a result that is not up-to-date). This is
commonly used in Banks, Retail store operations,
etc.

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 49

View Update
 Update on a view defined on a single table without any
aggregate functions
 Can be mapped to an update on underlying base table-

possible if the primary key is preserved in the view

 Update not permitted on aggregate views. E.g.,
UV2: UPDATE DEPT_INFO
SET Total_sal=100000
WHERE Dname=‘Research’;
cannot be processed because Total_sal is a computed value
in the view definition

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 51

View Update and Inline Views
 View involving joins
 Often not possible for DBMS to determine which
of the updates is intended
 Clause WITH CHECK OPTION
 Must be added at the end of the view definition if a
view is to be updated to make sure that tuples
being updated stay in the view
 In-line view
 Defined in the FROM clause of an SQL query (e.g.,
we saw its used in the WITH example)

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 51

Views as authorization mechanism
 SQL query authorization statements (GRANT and
REVOKE) are described in detail in Chapter 30
 Views can be used to hide certain attributes or
tuples from unauthorized users
 E.g., For a user who is only allowed to see
employee information for those who work for
department 5, he may only access the view
DEPT5EMP:
CREATE VIEW DEPT5EMP AS
SELECT *
FROM EMPLOYEE
WHERE Dno = 5;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 52

Schema Change Statements in SQL
 Schema evolution commands
 DBA may want to change the schema while the
database is operational
 Does not require recompilation of the database
schema

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 54

The DROP Command
 DROP command
 Used to drop named schema elements, such as
tables, domains, or constraint
 Drop behavior options:
 CASCADE and RESTRICT
 Example:
 DROP SCHEMA COMPANY CASCADE;
 This removes the schema and all its elements
including tables,views, constraints, etc.

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 55

The ALTER table command
 Alter table actions include:
 Adding or dropping a column (attribute)
 Changing a column definition
 Adding or dropping table constraints
 Example:
 ALTER TABLE COMPANY.EMPLOYEE ADD
COLUMN Job VARCHAR(12);

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 56

Adding and Dropping Constraints
 Change constraints specified on a table
 Add or drop a named constraint

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 57

Dropping Columns, Default Values
 To drop a column
 Choose either CASCADE or RESTRICT
 CASCADE would drop the column from views etc.
RESTRICT is possible if no views refer to it.
ALTER TABLE COMPANY.EMPLOYEE DROP COLUMN
Address CASCADE;
 Default values can be dropped and altered :
ALTER TABLE COMPANY.DEPARTMENT ALTER COLUMN Mgr_ssn
DROP DEFAULT;
ALTER TABLE COMPANY.DEPARTMENT ALTER COLUMN Mgr_ssn SET
DEFAULT ‘333445555’;

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 57

Table 7.2 Summary of SQL
Syntax

continued on next slide

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 59

Table 7.2 (continued)
Summary of SQL Syntax

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 60

Summary
 Complex SQL:
 Nested queries, joined tables (in the FROM
clause), outer joins, aggregate functions, grouping
 Handling semantic constraints with CREATE
ASSERTION and CREATE TRIGGER
 CREATE VIEW statement and materialization
strategies
 Schema Modification for the DBAs using ALTER
TABLE , ADD and DROP COLUMN, ALTER
CONSTRAINT etc.

Copyright © 2016 Ramez Elmasri and Shamkant B. Navathe Slide 7- 61