04 SQL

-4The Structured Query Language
SQL
SQL What is it ?
A powerful,
set oriented,
non-procedural
Non-proprietary
Language created to perform tasks such as update or
retrieve data from relational databases
First introduced by IBM in mid 70s
First commercial version by Oracle corporation in 79,
followed by IBM in 80
Available on ALL platforms from PC to mainframes
CS 53
Ramasesha Mudigere
. Contd. 2
SQL is ..
1. An interactive query language for ad-hoc queries
2. A database programming language
3. A combo of data definition, manipulation &
administration commands
4. Provides features, for data validation and transaction
support in a multi-user environment
5. Most appropriate language for networked database
servers
6. The flexible language that can be used for any data
storage and access
CS 53
Ramasesha Mudigere
SQL is ..
1. An interactive query language for ad-hoc queries
A
T
2. A database programming language e
A
h
D
t
f
r
o
3. A combo of data definition, manipulation
&l
o
f
o
r
t
administration commands nca
n
o
c
a
r
f
d and transaction
n
4. Provides features, for
data
validation
a
a
u
g
n
n
o
support in a multi
li -usernenvironment
ti
i
i
e
f
h
e
5. Most appropriate
language
for
networked
database
T
d
,
n
o
servers
i
t
a
l
u
p
6. The flexible
i language that can be used for any data
n
a
m and access
storage
CS 53
Ramasesha Mudigere
History Of SQL
In 1974, D. Chamberlin (IBM San Jose Laboratory)
defined language called Structured English Query
Language (SEQUEL)
A revised version, SEQUEL/2, was defined in 1976 but
name was subsequently changed to SQL for legal
reasons
IBM subsequently produced a prototype DBMS called
System R, based on SEQUEL/2
In late 70s, ORACLE appeared and was probably first
commercial RDBMS based on SQL
CS 53
Ramasesha Mudigere
SQL Standards
SQL-86 (A.k.a. SQL1)
issued by ANSI and ISO in 1986
SQL-89 - Minor revision like

intersection of SQL implementations of that time; (SQL
compliance was very easy for most)
Referential integrity
Bindings to standard programming languages
SQL-92 ( A.k.a. SQL2) -- Ratified in late 1992

Full-function RDBMS
Defines three levels of compliance - Entry, Intermediate, &

Full
The most widely used version in vogue even today
CS 53
Ramasesha Mudigere
. Contd.6
SQL Standards
SQL 1999 ( A.k.a. SQL3)
The current version of SQL standards

Over 5 times the size of SQL-92
Included OO features and stored procedures
Most DBMSs conform .. but some features are still not
implemented by some vendors.
SQL 2003 ( A.k.a. SQL4)

Proposed new Standard
XML Extensions and
SQL-1999 bug fixes
CS 53
Ramasesha Mudigere
SQL Standards
s s.
u
l
p
e
r
,
SQL 1999 ( A.k.a. SQL3)
u
s at
e
r
u y fe
t
a
The current version of SQL standards
e tar
f
2 ri e
9
p
Over 5 times the size of SQL-92
L
o
r
Q lp
S
ll , ci a
Included OO features and stored
procedures
a
t spe
o
n
f
i
&features are still not
Most DBMSs conform .. but
some
,
t
s
s ard
o
implemented by some
mvendors.
d
r
n
e sta
f
f
oSQL4)
r
SQL 2003 ( A.k.a.
s
e
t
m
a
l
e
t
s
m
y
Proposed
new
Standard
o
s
r
f
l
a
s
i
t
c
e
r
XML
Extensions
and
s
e
e
m tur
m
a
-1999 bug fixes
e
Co SQL
f
g
n
i
y
r
CS v53
Ramasesha Mudigere
8
a
SQL
Today,
300+ vendors of SQL products on various platforms
like
DB2, Oracle, Sybase, Informix, Microsoft SQL Server,
MSAccess, Ingress, MYSQL etc..
Most of them also have their own additional proprietary
extensions that are usually only used on their system
Has gone ( rather going ) through several revisions and
standardizations by ANSI
CS 53
Ramasesha Mudigere
SQL Commands
Can be Grouped as:
1. DML Component ) To manipulate data in the Tables
2. DDL Component ) To Add / Change / Delete Tables
3. Admin. Component ) To Control & manage access
permissions
4. Programming Component ) To access Database
in procedural way
5. Transaction Support ) To run transactions in a
multi-user environment
CS 53
Ramasesha Mudigere
10
1. SQL DML Component

Comprises FOUR (4 ) basic statements:
1. SELECT
To retrieve rows from tables
2. UPDATE
To modify the rows of tables
3. INSERT
To add new rows to tables
4. DELETE
CS 53
To remove rows from tables.

Ramasesha Mudigere
11
2. SQL DDL Component

Comprises THREE (3) basic statements:
1. CREATE
To create new table specification
2. DROP
To delete the tables
3. ALTER
To add or delete columns from a table, and
To change the specification on an existing
column
CS 53
Ramasesha Mudigere
12
3. SQL Admin Component

Comprises FOUR (4) basic statements:
1. CREATE
To create new users to access database
2. DROP
To delete the users
3. GRANT
To provide specific access permissions to users
4. REVOKE
To withdraw access permissions from users
CS 53
Ramasesha Mudigere
13
4. SQL Programming Component

1. DECLARE
To define CURSORS to access database
2. OPEN
To enable / activate CURSORS
3. FETCH
To get data rows based on CURSOR
4. CLOSE
To disable /deactivate the cursor
CS 53
Ramasesha Mudigere
14
5. SQL Transaction Component

1. BEGIN
To start a new transaction
2. COMMIT
To make the changes done by a a transaction
permanent
3. ROLLBACK
To undo the changes done by a transaction
4. SAVEPOINT
To create an intermediary commit point
CS 53
Ramasesha Mudigere
15
SQL Statements A Summary

DML
DDL
ADMIN
Support
SELECT
CREATE
CREATE
PGM
Support
TXN
Support
DECLARE
BEGIN
OPEN
UPDATE
DROP
DROP
INSERT
ALTER
GRANT
FETCH
ROLLBACK
REVOKE
CLOSE
SAVEPOINT
DELETE
CS 53
Ramasesha Mudigere
COMMIT
16
SQL Statements Coding Format

SQL is not Case Sensitive.
Exception:
Character values within quotation marks (Use the correct
case for these values.)
Example:
adams will not match with the stored value is Adams
Tabs and Indentations will promote readability.

Keywords cannot be split or abbreviated.
SQL Statements can be split across lines.
Clauses are placed in different lines, to promote
readability.
CS 53
Ramasesha Mudigere
17

1. SELECT
2. UPDATE
3. INSERT
4. DELETE
CS 53

Ramasesha Mudigere
18
SELECT
The most common and most used verb in SQL
The one ( and only one) basic command used to
retrieve data from column (s) of Table (s)
NOT to be confused with the SELECT operation of
Relational Algebra
So many options and flavor are available making it one
of the most versatile and powerful commands
Can vary from very simple to most Complex command
depending on options and parameters used
CS 53
Ramasesha Mudigere
19
An SQL SELECT Query

Has up to six clauses specified in the following order :
SELECT [ DISTINCT | ALL ] < attribute list >
FROM
< table list [alias] [, ...] >
[ WHERE
< condition > ]
[ GROUP BY
< grouping attribute(s) > ]
[ HAVING
< group condition > ]
[ ORDER BY
< attribute list> ]
Only the first two, SELECT and FROM, are mandatory

CS 53
Ramasesha Mudigere
20
Clauses of SELECT Statement

SELECT ) To List the columns (and expressions) that
should be returned from the query
FROM ) To Indicate the table(s) or view(s) from
which data will be obtained
WHERE ) To Indicate the conditions under which a
row will be included in the result
GROUP BY ) To Indicate categorization of results
HAVING ) To Indicate the conditions under which a
category (group) will be included
ORDER BY ) To Sort the result according to specified
criteria
CS 53
Ramasesha Mudigere
21
SQL statement
processing order
CS 53
Ramasesha Mudigere
22
SQL statement
processing order
CS 53
Ramasesha Mudigere
23
Simple SELECT command

1.
2.
3.
4.
SELECT * FROM EMPLOYEE

Lists all columns from table Employee with column name
( underlined ) as heading ;The output format looks something
like this
*
SELECT BDATE, ADDRESS FROM EMPLOYEE
Lists columns BDATE & ADDRESS for all Employees
SELECT DNO FROM EMPLOYEE
Lists dept number column from EMPLOYEE Table
SELECT DISTINCT DNO FROM EMPLOYEE
Lists only unique DNO ( 5, 4 and 1)

5.
CS 53
SELECT ALL DNO FROM EMPLOYEE

N
Ramasesha Mudigere
O/P Same As Q3
24
Simple SELECT with computed fields

Computed columns
Data that do not exist in the database Can be computed
using data in existing columns & arithmetic operators
SELECT SSN, LNAME, SALARY *12 FROM EMPLOYEE
Lists SSN, LNAME and SALARY *12 ( A computed figure ) for
all rows under the heading :
SSN
NAME
_SALARY * 12
CS 53
Ramasesha Mudigere
25
Simple SELECT with AS Clause

AS clause is used to name the column differently in the
output ( Not in the database )
AS clause is optional
SELECT SSN, LNAME AS NAME, SALARY *12 AS
ANNUAL_SALARY FROM EMPLOYEE
OR
SELECT SSN, LNAME NAME, SALARY *12

ANNUAL_SALARY FROM EMPLOYEE
Lists SSN, LNAME and SALARY *12 ( A computed figure ) for
all rows under the heading :
SSN
CS 53
NAME
Ramasesha Mudigere
ANNUAL_SALARY
26
CS 53
Ramasesha Mudigere
27
CS 53
Ramasesha Mudigere
B
28
SELECT Statement with WHERE Clause

SELECT FNAME , LNAME, SALARY FROM
EMPLOYEE WHERE DNO = 4
Lists the FNAME LNAME and SALARY from rows where
DNO = 4 ( 3 rows )
Comparison operators allowed are:
CS 53
Ramasesha Mudigere
29
SELECT Statement with compound Condition

Compound conditions
Formed by connecting two or more simple conditions
Uses AND, OR, and NOT operators
1. AND: all conditions must be true
2. OR: any one of conditions is true
3. NOT: reverses the truth of the original condition
The WHERE clause in SQL can include

compound Conditions to fine tune the selection
CS 53
Ramasesha Mudigere
30
SELECT Statement with compound

Condition
CS 53
Ramasesha Mudigere
31
SELECT Statement with compound

Condition
CS 53
Ramasesha Mudigere
32
Range Search Condition

1. BETWEEN operator
Used to search for a range of values
Inclusive values defining range are included
( BETWEEN 10 AND 20 will return 10 and 20 if available. )
Not an essential feature
Can arrive at same answer without it using AND
Does make certain SELECT commands simpler
Can be negated using NOT BETWEEN
CS 53
Ramasesha Mudigere
33
Range Search Condition - Example
CS 53
Ramasesha Mudigere
34
Pattern Matching Condition

LIKE operator
used when exact matches will not work
Can be used with a wildcard symbol
Can be negated using NOT LIKE
Wildcards vary depending on DBMS

% Matches one or more characters
_ Matches any one character
Can be used together and repeatedly
CS 53
Ramasesha Mudigere
35
Pattern Matching Condition

ENAME
SQL> SELECT ename
------2 FROM emp
JAMES
3 WHERE ename LIKE _A%; WARD
CS 53
Ramasesha Mudigere
36
Set Membership Condition

Use of IN
Provides concise way of phrasing certain conditions
Used to search for groups of values ( Replaces = )
List of values are in ( ) and comma separated
Advanced Advantage Can contain list of values from
another query
Can be negated by using NOT IN
List the customer number, last name, and first name

for every customer with a credit limit of $1,000, $1,500,
or $2,000
CS 53
Ramasesha Mudigere
37
Set Membership Condition

Use of IN
CS 53
Ramasesha Mudigere
38
ORDER BY Clause
Generally, the order of rows is immaterial to the DBMS
Results when displayed are in no defined order
Rows can displayed in the order in which they were
entered
However:
ORDER BY clause lists data in a specific order
Last part of SQL statement that controls order of records
returned
ASC Ascending Order (Default order )
DESC Descending Order
CS 53
Ramasesha Mudigere
39
ORDER BY Example
CS 53
Ramasesha Mudigere
40
ORDER BY Example
SQL> SELECT ename, job, deptno, hiredate
2 FROM
emp
3 ORDER BY hiredate DESC;
ENAME
JOB
DEPTNO
---------- --------- --------ADAMS
CLERK
20
SCOTT
ANALYST
20
MILLER
CLERK
10
JAMES
CLERK
30
FORD
ANALYST
20
KING
PRESIDENT
10
MARTIN
SALESMAN
30
...
1453 rows selected.
CS
Ramasesha Mudigere
HIREDATE
--------12-JAN-83
09-DEC-82
23-JAN-82
03-DEC-81
03-DEC-81
17-NOV-81
28-SEP-81
41
ORDER BY using multiple keys

When sorting more than one column
the more important column is called the major key
(or primary sort key)
the less important column is called the minor key
(or secondary sort key)
List keys in the order of importance in the ORDER BY

clause
CS 53
Ramasesha Mudigere
42
ORDER BY with multiple keys
CS 53
Ramasesha Mudigere
43
Handling NULL Values
signifies an unknown value or a value that does not exist.

Any arithmetic operation with NULL results in NULL
NULL values sometimes produce unexpected results
NULLS require special handling when coding
Use the SQL Operator IS NULL to test for NULL values
SQL> SELECT
2 FROM
3 WHERE
CS 53
ENAME
-------KING
ename, mgr
emp
mgr IS NULL;
MGR
---------
Ramasesha Mudigere
44
Aggregate operators in SQL

Also referred to as Group Functions
ISO standard defines five aggregate functions:
1. COUNT returns number of values in specified column
2. SUM
returns sum of values in specified column.
3. AVG
returns average of values in specified column.
4. MIN
returns smallest value in specified column.
5. MAX
returns largest value in specified column
All operators (Except count) used only on numeric data
ignoring NULL values, if any, are encountered
CS 53
Ramasesha Mudigere
45
COUNT
Returns the number of non-null occurrences of a column

Use of asterisk * Counts the rows
Use DISTICNCT to count duplicate values only once
SELECT COUNT(PART_NUMBER) FROM PART

WHERE ITEM_CLASS = HW;
CS 53
Ramasesha Mudigere
46
SUM, AVG, MAX and MIN

Work only on numeric fields
DISTICT can be used with SUM and AVG
Ignore null values ( But sometimes strange results are obtained)
CS 53
Ramasesha Mudigere
47
Summary of Group Functions
All of them Except count, operate on numeric fields only

COUNT alone
Can work on non-numeric as well as numeric fields
Can also use * as parameter to count number of rows
Null values are ignored during computation;

However strange results may occur sometimes with COUNT
Can be used in any combination

BUT
Can not be used together with direct column references
in the SELECT list (unless the SELECT statement
includes a GROUP BY clause)
CS 53
Ramasesha Mudigere
48
GROUP BY Clause
Intimately connected with Aggregates
Allows data to be:
grouped in a particular order and
compute statistics (aggregate functions) on these groups
that share some common characteristics

When used, the SELECT list must can contain
1. List of column names ( these names must appear in
GROUP BY Clause too) and
2. List of terms with aggregate operation
CS 53
Ramasesha Mudigere
49
GROUP BY Example
Consider the Table
CS 53
Find highest salary

Find highest salary in DepartmentNo = 5
Find DNO, and highest salary in each DNO
Ramasesha Mudigere
50
GROUP BY Example
SELECT DNO, MAX (SALARY) FROM EMPLYEE
GROUP BY DNO ORDER BY DNO
DNO
1
4
5
MAX (SALARY)
5500
4300
4000
If we need Max salary in each dept amongst female

employees only ?
SELECT DNO, MAX (SALARY) FROM EMPLYEE WHERE
SEX = F GROUP BY DNO ORDER BY DNO
CS 53
Ramasesha Mudigere
51
GROUP BY Another Example
CS 53
Ramasesha Mudigere
52
HAVING Clause
Designed for use with GROUP BY
Restricts groups that appear in final result table
Similar to WHERE, but WHERE filters individual rows whereas
HAVING filters groups
Column names in HAVING clause must also appear in the
GROUP BY list or be contained within an aggregate function
CS 53
Ramasesha Mudigere
53
HAVING Clause Another Example
CS 53
Ramasesha Mudigere
54
Clauses of SELECT Statement (Revisited)

SELECT ) To List the columns (and expressions) that
should be returned from the query
FROM ) To Indicate the table(s) or view(s) from
which data will be obtained
WHERE ) To Indicate the conditions under which a
row will be included in the result
GROUP BY ) To Indicate categorization of results
HAVING ) To Indicate the conditions under which a
category (group) will be included
ORDER BY ) To Sort the result according to specified
criteria
CS 53
Ramasesha Mudigere
55
Sub Query ( A. k. a. Nested Query)

Query within a Query
( SELECT embedded within another SELECT )
Can be used in WHERE and HAVING clauses of an
outer SELECT, where it is called a subquery or nested
query.
May also appear in INSERT, UPDATE, and DELETE
statements
OUTER Query
SELECT . FROM .
WHERE . ( SELECT .. FROM . WHERE ..)
CS 53
INNER
Ramasesha Mudigere
Query
56
Sub Query Example

Consider the Employee Table
List LNAME with salary higher than salary of Narayan
We need two Queries
SELECT SALARY FROM EMPLOYEE WHERE LNAME = Narayan
SALARY
3800
SELECT LNAME FROM EMPLOYEE WHERE SALARY > 3800
Instead we can Write a single nested query :

SELECT LNAME FROM EMPLOYEE WHERE SALARY > ( SELECT
SALARY FROM EMPLOYEE WHERE LANME = Narayan )
CS 53
Ramasesha Mudigere
57
EMPLOYEE TABLE
CS 53
Ramasesha Mudigere
58
Nested Query Further Example

Suppose we want to check Narayans salary is above
average in his company by finding names with salary
higher than the average salary of the company ?
AVG (SALARY) FROM EMPLOYEE)
Suppose we want to check Narayans salary is above

average in his Department alone
AVG (SALARY) FROM EMPLOYEE WHERE DNO = ( SELECT
DNO FROM EMPLOYEE WHERE LANME = Narayan )
CS 53
Ramasesha Mudigere
59
Nested Query More Examples

List the amount by which colleagues salaries higher
than that of Narayan
SELECT ( SALARY ( SELECT SALARY FROM EMPLOYEE
WHERE LANME = Narayan )) AS SALDIFF
FROM EMPLOYEE
WHERE SALARY > ( SELECT SALARY FROM EMPLOYEE
WHERE LANME = Narayan )
CS 53
Ramasesha Mudigere
60
Nested Query More Examples

List the Average of Excess salary of all those employees
who draw salary higher than the average salary of
Narayans department
SELECT AVG ( SALARY ( SELECT SALARY FROM EMPLOYEE
WHERE LANME = Narayan )) AS AVGEXCESS
FROM EMPLOYEE
WHERE SALARY > ( SELECT AVG (SALARY) FROM
EMPLOYEE WHERE DNO = ( SELECT DNO FROM
EMPLOYEE WHERE LANME = Narayan )
Sub-query returned single value output so far; How about

CS muti
53 -value return ?Ramasesha Mudigere
61
USE of IN in Nested Queries

Used to check the existence of a variable in outer query,
when inner query returns a set not a single value
Consider the Table Employee
List Name and salaries of all who work in Departments
with least one female employee
SELECT LNAME, SALARY FROM EMPLOYEE
WHERE DNO IN ( SELECT DISTINCT DNO FROM EMPLOYEE
WHERE SEX = F )
Inner Query Returns a set of DNO {4,5}

Outer query checks + condition for each member of set
CS 53
Ramasesha Mudigere
62
EMPLOYEE TABLE
CS 53
Ramasesha Mudigere
63
USE OF ANY & ALL

May be used with sub-queries that produce a single
column of numbers
With ALL, condition will only be true if it is satisfied by
all values produced by sub-query.
With ANY, condition will be true if it is satisfied by any
values produced by sub-query.
If subquery is empty, ALL returns true, ANY returns
false.
SOME may be used in place of ANY.
CS 53
Ramasesha Mudigere
64
ANY & ALL Example

Consider the Employee Table
List names of all employees whose salary is equal to
salary of any one who woks in dept 4
SELECT LNAME FROM EMPLOYEE WHERE SALARY = ANY
( SELECT DISTINCT SALARY FROM EMPLOYEE
WHERE DNO = 4 )
List names of all employees in other departments

whose salary is equal to salary of any one who woks
in dept 4
SELECT LNAME FROM EMPLOYEE WHERE Dno NOT = 4
AND SALARY = ANY ( SELECT DISTINCT SALARY FROM
EMPLOYEE WHERE DNO = 4 )
CS 53
Ramasesha Mudigere
65
EMPLOYEE TABLE
CS 53
Ramasesha Mudigere
66
Use OF EXISTS & NOT EXISTS

EXISTS and NOT EXISTS are for use only with
subqueries.
Produce a simple true/false result.
True if and only if there exists at least one row in result
table returned by subquery.
False if subquery returns an empty result table.
NOT EXISTS is the opposite of EXISTS.
CS 53
Ramasesha Mudigere
67
EXISTS Example
Consider Employee table
List all Names & department where at least one female
employee is working
SELECT LNAME, DNO FROM EMPLOYEE E
WHERE EXISTS ( SELECT * FROM EMPLOYEE M
WHERE E.DNO = M.DNO
= F )
AND M.SEX
M.
Note the use of variable from outer query in inner

query ( known as Correlated Query)
CS 53
Ramasesha Mudigere
68
SELECT With Multiple Tables

Used to retrieve data from two or more tables
Operation performed will be Joining of the tables
Since there is a possibility of referring to the same
column name in two different fields, the column names
need to be qualified with Table name as
Table name . Column name OR
as Table alias name . Column name
SELECT select_list
FROM table_1 AS T1, table_2 (,table_3)
WHERE T1. column join_operator table2. column
CS 53
Ramasesha Mudigere
69
SELECT with JOIN

Indicate in the SELECT clause all columns to display
List in the FROM clause all tables involved in the query
Give condition(s) in the WHERE clause to restrict the
data to be retrieved to only those rows that have
common values in matching columns
Use Alias , wherever it is necessary, to make the
reference unique
CS 53
Ramasesha Mudigere
70
How JOIN works in SQL

Form Cartesian product of the tables named in FROM
clause.
If there is a WHERE clause, apply the search condition to
each row of the product table, retaining those rows that
satisfy the condition.
For each remaining row, determine value of each item in
SELECT list to produce a single row in result table.
If DISTINCT has been specified, eliminate any duplicate
rows from the result table.
If there is an ORDER BY clause, sort result table as
required.
CS 53
Ramasesha Mudigere
71
Simple JOIN Example

Consider tables: Client & Viewing:
ID
CR76
CR56
CR74
CR62
ID
CR56
CR76
CR56
CSCR74
53
Name
Suresh
Sheela
Milind
Rajani
Phone
4563
7749
2275
9341
Pref
House
Flat
Flat
House
MaxRent
2000
1800
2000
2500
Property
PA14
PG4
PG4
PG36
Date Viewed
24-may-03
20-may-03
26-may-03
28-may-03
Ramasesha
Mudigere
Comments
Too Small
Too far
Looks OK
72
Simple JOIN Example

1. List names of all clients who have viewed a property
along with any comment supplied
SELECT C.ID, Name, property, comment FROM Client C, Viewing V
WHERE C.ID = V.ID;
Alternative ways to Specify JOINS

1. FROM Client C JOIN Viewing V ON C.ID = V.ID
2. FROM Client JOIN Viewing USING ID
3. FROM Client NATURAL JOIN Viewing
Notes
All the above 3 replace both FROM and WHERE in the SQL
First one produces a result with ID column appearing twice
CS 53
Ramasesha Mudigere
73
Relational Database Schema for Example

1
2
3
4
5
CS 53
Ramasesha Mudigere
74
Populated Database
2
5
4
6
CS 53
Ramasesha Mudigere
75
Multi Table SELECT Examples

Retrieve the name and address of all employees who
work for the 'Research' department.
Q1:
SELECT
FNAME, LNAME, ADDRESS
FROM EMPLOYEE, DEPARTMENT
WHERE
DNAME='Research' AND
DNUMBER=DNO
(DNAME='Research') is a selection condition

(corresponds to a SELECT operation in relational algebra)
(DNUMBER=DNO) is a join condition (corresponds to a
JOIN operation in relational algebra)
CS 53
Ramasesha Mudigere
76
SELF JOIN Example

For each employee, retrieve the employee's name, and
the name of his or her immediate supervisor.
SELECT
E.FNAME, E.LNAME, S.FNAME, S.LNAME
FROM
EMPLOYEE AS E S
WHERE
E.SUPERSSN=S.SSN
Note the use of on table employee as two different tables
for join with two different alias names E and S
We can think of E and S as two different copies of
EMPLOYEE, representing employees in role of
supervisees and supervisors respectively
This Process is known as Self Join

CS 53
Ramasesha Mudigere
77
OUTER JOIN
If one row of a joined table is unmatched, row is
omitted from result table of a JOIN.
Outer join operations retain rows that do not satisfy the
join condition.
Outer join can be
LEFT
Rows from left Table are retained
RIGHT
Rows from the right table are retained or
FULL
Rows from the Both tables are retained
CS 53
Ramasesha Mudigere
78
Consider tables above:

The (inner) join of these two tables: SELECT b.*, p.* FROM
Branch1 b, PropertyForRent1 p WHERE b.bCity = p.pCity;
Results in two rows
To include unmatched rows in result table, use an Outer join

CS 53
Ramasesha Mudigere
79
Outer JOIN Example - LEFT

List branches and properties that are in same city along
with any unmatched branches
SELECT b.*, p.* FROM Branch1 b LEFT JOIN PropertyForRent1 p
ON b.bCity = p.pCity;
Includes those rows of first (left) table unmatched with

rows from second (right) table.
Columns from second table are filled with NULLs
CS 53
Ramasesha Mudigere
80
Outer JOIN Example - RIGHT

List branches and properties that are in same city along
with any unmatched Properties
SELECT b.*, p.* FROM Branch1 b RIGHT JOIN PropertyForRent1
p ON b.bCity = p.pCity;
Includes those rows of second (right) table unmatched

with rows from first (left) table.
Columns from First table are filled with NULLs
CS 53
Ramasesha Mudigere
81
Outer JOIN Example - FULL

List branches and properties in same city and any
unmatched branches or properties
SELECT b.*, p.* FROM Branch1 b FULL JOIN PropertyForRent1 p
ON b.bCity = p.pCity;
Includes rows that are unmatched in both tables.

Unmatched columns are filled with NULLs.
CS 53
Ramasesha Mudigere
82
Set Operations with SELECT

Supports UNION, INTERSECTION and DIFFERENCE
to combine results of two or more queries whose
results are union compatible into a single result table
Format of set operator clause in each case is:
op [ALL] [CORRESPONDING [BY {column1 [, ...]}]]
If CORRESPONDING BY specified, set operation
performed on the named column(s).
If CORRESPONDING specified but not BY clause,
operation performed on common columns.
If ALL specified, result can include duplicate rows
CS 53
Ramasesha Mudigere
83
Set Operation Examples - UNION

Consider the following tables:
Branch ( Branchno, street, city, pin)
Property (Propid, street, city, type, rent, ownerid, branchandling)
1. List all cities with either a branch office or a property

SELECT city FROM Branch WHERE city IS NOT NULL)
UNION
SELECT city FROM Property WHERE city IS NOT NULL
OR
SELECT * FROM Branch WHERE city IS NOT NULL
UNION CORRESPONDING BY city
SELECT * FROM Property WHERE city IS NOT NULL
CS 53
Ramasesha Mudigere
84
Set Operation Examples - INTERSECT

2. List all cities with both branch office AND a property
SELECT city FROM Branch
INTERSECT
SELECT city FROM Property
OR
SELECT * FROM Branch
INTERSECT CORRESPONDING BY city
SELECT * FROM Property
CS 53
Ramasesha Mudigere
85
Set Operation Examples - EXCEPT

3. List all cities with branch office but no property
SELECT city FROM Branch
EXCEPT
SELECT city FROM Property
OR
SELECT * FROM Branch
EXCEPT CORRESPONDING BY city

SELECT * FROM Property
CS 53
Ramasesha Mudigere
86
Summary of SQL Queries (cont.)

The SELECT-clause lists the attributes or functions to be
retrieved
The FROM-clause specifies all relations (or aliases) needed in
the query but not those needed in nested queries
The WHERE-clause specifies the conditions for selection and join
of tuples from the relations specified in the FROM-clause
GROUP BY specifies grouping attributes
HAVING specifies a condition for selection of groups
ORDER BY specifies an order for displaying the result of a query
A query is evaluated by first applying the WHERE-clause, then
GROUP BY and HAVING, and finally the SELECT-clause
CS 53
Ramasesha Mudigere
87

1. SELECT
2. UPDATE
3. INSERT
4. DELETE
CS 53

Ramasesha Mudigere
88
UPDATE Command of SQL

Used to modify existing rows with the UPDATE
statement
Can Update more than one row at a time, if required
The format of the UPDATE command is:
UPDATE TableName
SET colName1 = dataValue1 [, colName2 = dataValue2...]
[WHERE searchCondition]
SET clause specifies names of one or more columns

along with new compatible values that are to be updated
WHERE limits rows updated
If omitted all rows are Updated
CS 53
Ramasesha Mudigere
89
UPDATE Example
1. Give all staff a 3% pay increase.
UPDATE Staff SET salary = salary*1.03;
2. Give all Managers a 5% pay increase.

UPDATE Staff SET salary = salary*1.05
WHERE position = Manager;
3. Promote XYZ (staffNo=SG14) to Manager & change

his salary to 55,000
UPDATE Staff
SET position = Manager, salary = 55000
CS 53
WHERE staffNo = SG14
Ramasesha Mudigere
90

1. SELECT
2. UPDATE
3. INSERT
4. DELETE
CS 53

Ramasesha Mudigere
91
INSERT Command of SQL

Add a new row to a table
INSERT INTO TableName [ (colList) ] VALUES (dataValueList)
columnList is optional; if omitted, SQL assumes a list of all

columns in their original CREATE TABLE order.
Any columns omitted must have been declared as NULL when
table was created, unless DEFAULT was specified when
creating column
dataValueList must match colList as follows:
number of items in each list must be same;
must be direct correspondence in position of items in two lists;
data type of each item in dataValueList must be compatible with
data type of corresponding column.
CS 53
Ramasesha Mudigere
92
INSERT Example
SQL> INSERT INTO department (deptno, dname, loc)
2 VALUES
(50, 'DEVELOPMENT', 'DETROIT');
1 row created.
Implicit method: Omit the column from the column list

SQL> INSERT INTO department (deptno, dname )
2 VALUES
(60, 'MIS');
1 row created.
Explicit method: Specify the NULL keyword

SQL> INSERT INTO department
2 VALUES
(70, 'FINANCE', NULL);
1 row created.
CS 53
Ramasesha Mudigere
93
INSERT with a sub-query
Allows for multiple rows to be copied from one or more

tables to another
Do not use the VALUES clause .
Match the number of columns in the INSERT clause to
those in the subquery
SQL> INSERT INTO managers (id, name, salary,
hiredate)
2
SELECT empno, ename, sal, hiredate
3
FROM employee
4
WHERE job = 'MANAGER'
3 rows created.
CS 53
Ramasesha Mudigere
94

1. SELECT
2. UPDATE
3. INSERT
4. DELETE
CS 53

Ramasesha Mudigere
95
DELETE Command of SQL

Used to remove existing rows from a table
DELETE FROM TableName [ WHERE searchCondition ]
SearchCondition is optional;
If specified, only the rows that satisfy condition are deleted
if omitted, all rows are deleted ( But not the table itself !)
1. SQL> DELETE FROM department
2 WHERE
dname = 'DEVELOPMENT'
1 row deleted.
2. SQL> DELETE FROM department;
4 rows deleted.
CS 53
Ramasesha Mudigere
96

1. SELECT
2. UPDATE
3. INSERT
4. DELETE
CS 53

Ramasesha Mudigere
97

1. CREATE
2. DROP
3. ALTER
column
CS 53
Ramasesha Mudigere
98
SQL Data Types
1. Boolean ) Consists of variables which can take truth values
of TRUE, FALSE & UNKNOWN (NULL)

2. Character ) Consists of sequence of characters ( a string)
from vendor defined set ( ASCII or EBCDIC)
3. Numeric ) To define numbers ( both integer and real ) based
on pattern or storage allocation
4. BIT ) Consists of sequence of bits ( 0 / 1) ( bit string)

5. Datetime ) To define points in time to a certain degree of
accuracy
6. Interval ) Used to represent periods of time
7. Large Objects ( LOBs) ) Used to store huge amounts of

binary/character data
CS 53
Ramasesha Mudigere
99
2. SQL Data Types Character types

Two sub types are supported namely
1. Fixed length CHAR (n) where n is number of Characters
2. Variable length VARCHAR (n) or CHAR VARYING (n)
where n is the max size of the data item
When specifying a literal string value, the string is

placed between single Quotes ( E.g. Example)
When the character string stored in a field is less than
the size defined, Fixed length CHAR fields are blank
filled to the right & these blanks are ignored in
comparison
CS 53
Ramasesha Mudigere
100
3. SQL Data Types Numeric types

Two sub-categories namely
1. Approximate Numeric which describes the type &
2. Exact Numeric which describes format of the numbers
2.1 Approximate Numeric has following types
CS 53
INTEGER or INT
SMALLINT
FLOAT ( Precision)
or REAL
DOUBLE PREISION
Precision on FLOAT defines the precision of mantissa
Precision & size of REAL & DOUBLE PRECISION is
implementation dependent
Ramasesha Mudigere
101
. Contd.
3. SQL Data Types Numeric types

Two sub-categories namely
1. Approximate Numeric which describes the type &
2. Exact Numeric which describes format of the numbers
2.2 Exact Numeric
Used for declaring formatted numbers

NUMERIC ( p, s ) or DECIMAL ( p, s) where
p represents the TOTAL number of digits ( precision) and
S represents the number of digits to the right of decimal
point ( scale) ( 0 < or = s < or = p & default is 0)
Example : SALARY TYPE NUMERIC (5,2) can hold a
maximum value of 999.99 & min value of 999.99
CS 53
Ramasesha Mudigere
102
4. SQL Data types Bit String Type

Used to store strings of bits that can take value 0 or 1
Two types are supported
BIT (n) where n specifies the length of the bit string
BIT VARYING (n) specifies a variable length string of
maximum size n
Literal bit strings are specified by placing the string
between single quotes but preceded by B ( like fo
example B101101
1.
2.
If the string size is a multiple of even hexa-decimal

notation can be used with preceding character X
Data is right justified and zero filled to the left
CS 53
Ramasesha Mudigere
103
5. SQL Data types Datetime Type

Used to define points in time to a certain degree of
accuracy
E.g. dates, times, and times of a day
ISO standard subdivides this data type to YEAR, MONTH,

DAY, HOUR, MINUTE, SECOND, TIMEZONE_ HOUR &
TIMEZONE_MINUTE
1.
2.
3.
Three types of Datetime data type is supported

DATE
TIME [ timePrecision] WITH TIME ZONE]
TIMESTAMP [ timePrecision] WITH TIME ZONE]
CS 53
Ramasesha Mudigere
104
. Contd.

5.1 DATE
Used to store calendar dates
Has TEN positions with components as YEAR MONTH and DAY
in the form YYYYMMDD
Literal values can be mentioned as DATE2006-09-27 and
comparison performed directly
5.2 TIME [ timePrecision] WITH TIME ZONE]

Used to store time with at least 8 positions with components
HOUR, MINUTE and SECOND in the form HH:MM:SS
timeprecision is the number of the decimal places of accuracy
to which the second field is kept ( Max vendor dependent) and
it defaults to 0
CS 53
Ramasesha Mudigere
CS 53
Ramasesha Mudigere
105
. Contd.

5.2 TIME [ timePrecision] WITH TIME ZONE]
WITH TIME ZONE keyword includes additional six positions
for specifying the displacement from the standard universal
time zone
Literal values can be mentioned as TIME09:22:15 and
comparison performed directly
5.3 TIMESTAMP
Includes both the DATE and TIME fields plus a minimum of
six positions for decimal fractions of seconds and optional with
TIME ZONE Qualifier
Literal values are placed by single-quoted-strings preceded by
TIMESTAMP ( E.g. TIMESTAMP 2006-09-1115:35:648302)
CS 53
Ramasesha Mudigere
106
6. SQL Data types Interval Type

Used to represent periods of time ( Durations)
Consists of contiguous subset of fields YEAR, MONTH,
DAY, HOUR, MINUTE, SECOND
Two types of INTERVAL fields

1. Year-Month intervals &
2. Day-time Intervals
Generally used in Temporal databases
Format
CS 53
INTERVAL [{startfield] TO [Endfield}] or [singletimefield]

Ramasesha Mudigere
107
7. SQL Data types LOBs

Used to store large size objects which are generally
not kept as a part of data base (E.g. (photos, videos,
CAD files, etc.)
1. blob: binary large object
A large collection of un-interpreted binary data (whose

interpretation is left to an application outside of the
database system)
2. clob: character large object
a large collection of character data
When a query returns a large object, a pointer is

returned rather than the large object itself.
CS 53
Ramasesha Mudigere
108
8. User Defined data Types

SQL 3 supports user Defined data types (UDT)
A specific domain can be declared and the domain
name used with the attribute Specification
CREATE DOMAIN AGE_TYPE AS numeric (2)
CHECK(VALUE BETWEEN 16 AND 58);
The domain may later be used as data type in defining
the column type as AGE_TYPE
CS 53
Ramasesha Mudigere
109
ISO Standard SQL Data Types
CS 53
Ramasesha Mudigere
110
Data Types Supported in SQL 92 ( ORACLE)

Data type
Description
VARCHAR2(size)
Variable-length character data
CHAR(size)
Fixed-length character data
NUMBER(p,s)
Variable-length numeric data
DATE
Date and time values
LONG
Variable-length character data up to 2 GB
CLOB
Single-byte character data up to 4 GB
RAW and LONG RAW

BLOB
BFILE
CS 53
Raw binary data
Binary data up to 4 gigabytes

Binary data stored in an external file; up to 4 GB
Ramasesha Mudigere
111
Integrity Enhancement Features

SQL Standards provide for Integrity Control using
which we can define constraints, that protect database
from becoming inconsistent
FIVE types of Integrity constraints are supported
1. Required data.
2. Domain constraints.
3. Entity integrity.
4. Referential integrity.
5. Enterprise constraints
CS 53
Ramasesha Mudigere
112
. Contd.
Integrity Enhancement Features (Contd.)

1. Required data
A mandatory column that must have data in it can be

Specified using NOT NULL clause as:
position
VARCHAR(10)
NOT NULL
All primary key fields MUST have this clause
2. Domain Constraint
CS 53
Set of legal values for the variable can be specified

Gender CHAR NOT NULL CHECK (Gender IN (M, F))
Can also create a specific domain with CHECK
constraints that can be used with several columns in the
entire database
Ramasesha Mudigere
113
. Contd.

3. Entity Integrity
Specified at the table level
Primary key of a table must contain a unique, non-null

value for each row.
Can have ONLY ONE primary key definition per table (

Could be made up of multiple columns though)
Can still ensure uniqueness for alternate keys using

UNIQUE clause as:
CS 53
UNIQUE ( Tel-No)
Ramasesha Mudigere
114
. Contd.

4. Referential Integrity
Achieved using FOREIGN KEY (FK) Definition

(FK is column or set of columns that links each row in
child table containing foreign FK to row of parent table
containing matching PK. ) For example
FOREIGN KEY ( branchNo ) REFERENCES Branch
CS 53
Referential integrity means that, if FK contains a value,

that value must refer to existing row in parent table
Any INSERT that attempts to create FK value in child

table without matching candidate key value in parent is
rejected.
Ramasesha Mudigere
115
. Contd.

4. Referential Integrity ( Contd.)
Referential Action
CS 53
Action taken when attempts to update/delete a candidate

key value in parent table with matching rows in child is
dependent on referential action
Specified using ON UPDATE and ON DELETE subclauses of FOREIGN KEY clause of CREATE Statement
Possible Options are
1. CASCADE
2. SET NULL
3. SET DEFAULT &
4. NO ACTION
Ramasesha Mudigere
116
. Contd.

Referential Action
CASCADE ) Automatically delete the matching rows in the
child table recursively
SET NULL ) Set the foreign Key column in child table to
NULL ( Valid only if that column is not
declared as NOT NULL)
SET DEFAULT ) Set the foreign Key column in child table

to DEFAULT Value ( Valid only if that
column is declared with a DEFAULT
option)
NO ACTION ) Reject the command (Default Action)

CS 53
Ramasesha Mudigere
117

5. Enterprise Constraints
A set of rules that are application specific that can be set
Done Using CHECK / UNIQUE clauses in CREATE and

ALTER TABLE Statements of SQL
No manager can manage more than 100 properties.
CREATE ASSERTION StaffNotHandlingTooMuch
CHECK (NOT EXISTS (SELECT staffNo

FROM PropertyForRent
GROUP BY staffNo
HAVING COUNT(*) > 100))
CS 53
Ramasesha Mudigere
118
CREATE Command of SQL

Creates a table with one or more columns of the
specified data Type.
With any column
Specifying NOT NULL system rejects any attempt to insert
a null in the column.
UNIQUE specification rejects duplicates in the column
CHECK clause can be specified to restrict allowed values
A DEFAULT value for the column Can be specified
FOREIGN KEY clause specifies FK along with the
referential action
Primary keys can be specified for each table

CS 53
Ramasesha Mudigere
119
CREATE Command Syntax

CREATE TABLE TableName
{(colName dataType [NOT NULL] [UNIQUE]
[DEFAULT defaultOption]
[CHECK searchCondition] [,...]}
[ CONSTRAINT] [PRIMARY KEY (listOfColumns),]
{[UNIQUE (listOfColumns),] [,]}
{[FOREIGN KEY (listOfFKColumns)
REFERENCES ParentTableName [(listOfCKColumns)],
[ON UPDATE referentialAction]
[ON DELETE referentialAction ]] [,]}
{[CHECK (searchCondition)] [,] })
CS 53
Ramasesha Mudigere
120
Defining attributes and their data types
CS 53
Ramasesha Mudigere
121
CREATE Command Example
CS 53
Ramasesha Mudigere
122
CS 53
Ramasesha Mudigere
123
CS 53
Ramasesha Mudigere
124
Assertions
A predicate expressing a condition that we wish the
database always to satisfy
An assertion in SQL takes the form
create assertion <assertion-name> check
<predicate>
When an assertion is made, the system tests it for
validity, and tests it again on every update that may
violate the assertion
May introduce a significant amount of overhead; hence
assertions should be used with great care.
CS 53
Ramasesha Mudigere
125
Assertions
Every loan has at least one borrower who maintains an
account with a minimum balance or $1000.00
create assertion balance_constraint check
(not exists (
select *
from loan
where not exists (
select * from borrower, depositor, account
where loan.loan_num = borrower.loan_num
and borrower.cust_name = depositor.cust_name
and depositor.acct_num = account.acct_num
and account.balance >= 1000)))
CS 53
Ramasesha Mudigere
126

1. CREATE
2. DROP
3. ALTER
column
CS 53
Ramasesha Mudigere
127
DROP Command
Removes named table and all rows within it.
Format :
DROP TABLE TableName [RESTRICT | CASCADE]
e.g. DROP TABLE PropertyForRent;
With RESTRICT, if any other objects depend for their

existence on continued existence of this table, SQL
does not allow request.
With CASCADE, SQL drops all dependent objects (and
objects dependent on these objects).
CS 53
Ramasesha Mudigere
128

1. CREATE
2. DROP
3. ALTER
column
CS 53
Ramasesha Mudigere
129
ALTER Command
Can be used for following functions
1. Add a new column to a table.
2. Drop a column from a table.
3. Add a new table constraint.
4. Drop a table constraint.
5. Set a default for a column.
6. Drop a default for a column.
CS 53
Ramasesha Mudigere
130
ALTER command Format

ALTER TABLE TableName
[ADD [COLUMN] Colname, datatype [NOT NULL]., [ UNIQUE]
[ DEFAULT defaultOption ] [CHECK searchCondition]
[ DROP [COLUMN} colname [ RESTRICT [ CASCADE ]]
[ADD [ CONSTRAINT [ ConstraintName]]
[DROP [ CONSTRAINT [ ConstraintName]]
[ALTER [COLUMN] SET DEFAULT defaultOption]
[ALTER [COLUMN] DROP DEFAULT defaultOption]
CS 53
Ramasesha Mudigere
131
ALTER Examples
Change Staff table by removing default of Assistant

for position column and setting default for sex column
to female (F).
ALTER TABLE Staff ALTER position DROP DEFAULT;
ALTER TABLE Staff ALTER sex SET DEFAULT F;
Remove constraint from Property that staff not allowed

to handle more than 100 properties at time
ALTER TABLE Property
DROP CONSTRAINT StaffNotHandlingTooMuch;
Add new column to Client table

ALTER TABLE Client ADD prefNoRooms NUMERIC (2);
CS 53
Ramasesha Mudigere
132
Steps in table creation:

1. Identify data types for attributes
2. Identify columns that can and cannot be null
3. Identify columns that must be unique (candidate keys)
4. Identify primary key-foreign key mates
5. Determine default values
6. Identify constraints on columns (domain specifications)
7. Create the table and associated indexes
CS 53
Ramasesha Mudigere
133
VIEWS in RDBMS
A virtual table created dynamically with one or more
relational operations on base relations
Virtual relation that does not necessarily exist in the
DB, but is produced upon request, at time of request.
Contents of a view are defined as a query on one or
more base relations or other views
Allows full query operations ( With some restricitons)
Allows for limited update operations (since the table
may not physically be stored)
CS 53
Ramasesha Mudigere
134
. Contd.
VIEWS in RDBMS
With view resolution,
any operations on view are automatically translated into

operations on relations from which it is derived.
With view materialization

the view is stored as a temporary table, which is maintained
as the underlying base tables are updated
assumption: other queries on the view will follow
concerns: maintaining correspondence between the base
table and the view when the base table is updated
strategy: incremental update
Generally used in SQL procedural programming
environments
CS 53
Ramasesha Mudigere
135
. Contd.
Views Advantages & Disadvantages
Advantages:
Simplify query commands

Assist with data security (but don't rely on views for
security, there are more important security measures)
Enhance programming productivity
Contain most current base table data
Use little storage space
Provide customized view for user
Establish physical data independence
Disadvantages
Use processing time each time view is referenced
May or may not be directly updateable
CS 53
Ramasesha Mudigere
136
Creation of VIEWS
Using CREATE Command with following formant:

CREATE VIEW ViewName [ (newColumnName [,...]) ]
AS sub-select [WITH [CASCADED | LOCAL] CHECK OPTION]
Can assign a name to each column in view.

If list of column names is specified, it must have same number
of items as number of columns produced by sub-select.
((A.k.a. defining query ; it cant have ORDER BY)
If omitted, each column takes name of corresponding column
in sub-select
Need SELECT privilege on all tables referenced in subselect
and USAGE privilege on any domains used in referenced
columns
CS 53
Ramasesha Mudigere
137
CREATE VIEW ( horizontal) Example

Create view so that manager at branch B003 can only
see details for staff who work in his or her office.
CREATE VIEW Manager3Staff
AS SELECT * FROM Staff WHERE branchNo = B003;
CS 53
Ramasesha Mudigere
138
CREATE VIEW ( Vertical) Example

Create view of staff details at branch B003 excluding
salaries
CREATE VIEW Staff3
AS SELECT staffNo, fName, lName, position, sex
FROM Staff WHERE branchNo = B003;
CS 53
Ramasesha Mudigere
139
Grouped and Joined Views

A complex view that contains group functions to
display values from two tables
SQL> CREATE VIEW
2
3
(name, minsal, maxsal, avgsal)

AS SELECT d.dname, MIN(e.sal), MAX(e.sal),
AVG(e.sal)
FROM
WHERE
GROUP BY
View created.
CS 53
dept_sum_vu
employee e, department d
e.deptno = d.deptno
d.dname;
Ramasesha Mudigere
140
DML Operations on a View

Cannot update / delete a row if the view contains any
of the following:
Group functions
A GROUP BY clause
The DISTINCT keyword
Columns defined by expressions

Can not add data through INSERT INTO view if:
The view contains any of the conditions mentioned above
There are NOT NULL columns in the base tables that are
not selected by the view
CS 53
Ramasesha Mudigere
141
WITH CHECK OPTION on View CREATE

Must be added to the definition of a view if the view is to
be updated
Rows exist in a view because they satisfy WHERE
condition of defining query.
If a row changes and no longer satisfies condition, it
disappears from the view.
New rows appear within view when insert/update on view
cause them to satisfy WHERE condition.
Rows that enter or leave a view are called migrating
rows.
WITH CHECK OPTION prohibits a row migrating out of the
view
CS 53
Ramasesha Mudigere
142
WITH CHECK OPTION - Example

Consider the view defined as
CREATE VIEW Manager3Staff
AS SELECT * FROM Staff
WHERE branchNo = B003
WITH CHECK OPTION;
Cannot update branch number of row B003 to B002 as
this would cause row to migrate from view.
Also cannot insert a row into view with a branch
number that does not equal B003.
CS 53
Ramasesha Mudigere
143
Removing Views
Done using DROP Command
DROP VIEW ViewName [RESTRICT | CASCADE]
Causes definition of view to be deleted from database.
For example:
DROP VIEW Manager3Staff;
NOTE:
Unlike the dropping of tables, the data rows still remain
even after View is dropped, since data rows belong to
base tables and NOT views
CS 53
Ramasesha Mudigere
144
INDEX
Used to speed up the retrieval of rows by using a pointer

Reduces disk I/O by using rapid path access method to
locate the data quickly
Independent of the table it indexes
Automatically used and maintained by the RDBMS
Index is created
Automatically
A unique index is created automatically when you define a
PRIMARY KEY or UNIQUE key constraint in a table definition.
Manually
CS 53
Users can create non-unique indexes on columns to speed up

access time to the rows
Ramasesha Mudigere
145
Creating an Index - Example

The Syntax is :
CREATE INDEX index
ON table (column[, column]...);
Improve the speed of query access on the ENAME
column in the EMP table
SQL> CREATE INDEX
2 ON
Index created.
CS 53
emp_ename_idx
employee(ename);
Ramasesha Mudigere
146
Removing the Index

DROP command is used to remove the index from DB
The Syntax is :
DROP INDEX index
Example:
Remove the EMP_ENAME_IDX index from the data
dictionary
SQL> DROP INDEX emp_ename_idx;
Index dropped.
CS 53
Ramasesha Mudigere
147

1. CREATE
2. DROP
3. ALTER
column
CS 53
Ramasesha Mudigere
148

1. CREATE
2. DROP
To delete the users
3. GRANT
4. REVOKE
CS 53
Ramasesha Mudigere
149
CREATE & DROP Users

The DBA creates users by using the CREATE USER
statement CREATE USER
user
Format : IDENTIFIED BY password;
Example:
SQL> CREATE USER scott
2 IDENTIFIED BY tiger;
User created.
Users are Removed Using DROP command

SQL> DROP USER
CS 53
User dropped.
scott
Ramasesha Mudigere
150

1. CREATE
2. DROP
To delete the users
3. GRANT
4. REVOKE
CS 53
Ramasesha Mudigere
151
GRANT Command
Used to provide privileges to users

Privileges:
Actions user permitted to carry out on given base table or view:

SELECT
Retrieve data from a table.
INSERT Insert new rows into a table.
UPDATE
Modify rows of data in a table.
DELETE
Delete rows of data from a table.
REFERENCESReference columns of named table in
integrity constraints.
USAGE Use domains, collations, character sets, and
translations.
Can restrict INSERT/UPDATE/REFERENCES to named columns.
CS 53
Ramasesha Mudigere
152
GRANT Command
GRANT {PrivilegeList | ALL PRIVILEGES}
ON ObjectName TO {AuthorizationIdList | PUBLIC}
[WITH GRANT OPTION]
PrivilegeList consists of one or more of privileges
separated by commas.
ALL PRIVILEGES grants all privileges to a user
GRANT Can be executed by the owner of the table
PUBLIC allows access to be granted to all present and
future authorized users
WITH GRANT OPTION allows privileges to be passed
on
CS 53
Ramasesha Mudigere
153
GRANT Example
Give Scott full privileges to Staff table.
GRANT ALL PRIVILEGES
ON Staff
TO Scott WITH GRANT OPTION;
Give users Scott and Narayan SELECT and UPDATE
on column salary of Staff.
GRANT SELECT, UPDATE (salary)
ON Staff
TO Scott, Narayan ;
CS 53
Ramasesha Mudigere
154
Concept of ROLE
Users
Manager
Privileges
Allocating privileges
Without a role
CS 53
Ramasesha Mudigere
Allocating privileges
with a role
155
GRANTing with ROLEs - Example

SQL>
SQL>
Role
Role
CREATE
CREATE ROLE
ROLE manager;
manager;
created.
created.
SQL>
SQL> GRANT
GRANT create
create table,
table, create
create view
view
22
to
to manager;
manager;
Grant
Grant succeeded.
succeeded.
SQL>
SQL> GRANT
GRANT manager
manager to
to BLAKE,
BLAKE, CLARK;
CLARK;
Grant
Grant succeeded.
succeeded.
CS 53
Ramasesha Mudigere
156

1. CREATE
2. DROP
To delete the users
3. GRANT
4. REVOKE
CS 53
Ramasesha Mudigere
157
REVOKE Command
Takes away privileges granted with GRANT
The format is
REVOKE [GRANT OPTION FOR]
{PrivilegeList | ALL PRIVILEGES}
ON ObjectName
FROM {AuthorizationIdList | PUBLIC}
[RESTRICT | CASCADE]
ALL PRIVILEGES refers to all privileges granted to a
user by user revoking privileges.
CS 53
Ramasesha Mudigere
158
REVOKE Command Options

GRANT OPTION FOR allows privileges passed on via
WITH GRANT OPTION of GRANT to be revoked
separately from the privileges themselves.
Privileges granted to this user by other users are not
affected.
Example:
SQL> REVOKE select, insert
2 ON department
3 FROM
scott;
Revoke succeeded.
CS 53
Ramasesha Mudigere
159

1. CREATE
2. DROP
To delete the users
3. GRANT
4. REVOKE
CS 53
Ramasesha Mudigere
160
Database Programming
Objective: ) To access a database from an application
program (as opposed to interactive interfaces)

Reason ) An interactive I/F is convenient but not sufficient;
) Many DB operations are made thru application
programs (nowadays thru web applications
Solution: ( Two fold)

1. Embedded SQL ) 1.Database commands are embedded in a
general-purpose programming languages
) 2. Augment SQL with program constructs
E.g. PL/SQL , TRANSACT_SQL etc.
2. API ) Library of database functions available to the host
language for database calls ( ODBC, JDBC, CLI )
CS 53
Ramasesha Mudigere
161
Database Programming Issues

The basic issue is the Incompatibilities between a host
programming language and the database model
(Impedance Mismatch) like.,
1. Type mismatch and incompatibilities; requires a new
binding for each language
2. set vs. record-at-a-time processing
Need special iterators to loop over query results and
manipulate individual values
CS 53
Ramasesha Mudigere
162
1. Type mismatch problem
Solution
Try and match the DBMS data types to the
corresponding closest match in the language like:
CHAR datatype to CHAR in C and PIC X field in COBOL
NUMERIC to Char in C ( with external to care to store only
numbers) and PIC 9 fields in COBOL
Define a mechanism to declare variables to be used as

interface between DB attributes and program variables
An EXEC DECLARE SECTION
Define an elaborate error interface to report errors

back to the programs in case of type mismatch after
each SQL execution in the program
CS 53
Ramasesha Mudigere
163
2. Set vs. record-at-a-time processing

Solution: cursors
Functions as follows:
Open (a table or a result table): position the cursor just
before the first row
Get next: move the cursor to the next row and return that
row; raise a flag if there is no more next row
Close: clean up and release DBMS resources
Found in virtually every database language/API (with

slightly different syntaxes)
Some support more cursor positioning and movement
options, modification at the current cursor position, etc.
CS 53
Ramasesha Mudigere
164
Steps IN DB Programming
How it all works ?
1. Client program opens a connection to the database server
2. Client program submits queries to and/or updates the DB
3. When database access is no longer needed, client program
terminates the connection
What needs to be done ?

1. Declare variables to be used as interfaces in the program
1. Data items ( A.k.a. Host Language variables)
2. Error handling & Communication
2. Embed SQLs where needed in the program

3. Devise an error Handling Strategy
CS 53
Ramasesha Mudigere
165
Embedding SQL
Embedded SQL starts with identifier,
usually EXEC SQL [ @SQL( in MUMPS].
Ends with terminator dependent on host language:
Ada, C, and PL/1: terminator is semicolon (;)

COBOL: terminator is END-EXEC
FORTRAN: ends when no more continuation lines.
Embedded SQL can appear anywhere an executable

host language statement can appear
Will be processed by the pre-processor before
language compiler is invoked
CS 53
Ramasesha Mudigere
166
Embedded Program Preparation

Pgm
Editor
Save
Source
Module
precompiler
Pre-compile
Object
Module
ORACLE
Run Time
Library
CS 53
Compile
Modified
Source
Module
Compiler
Linker
Ramasesha Mudigere
Link
Application
Program
167
CS 53
Ramasesha Mudigere
168
Host language Variables

Used in embedded SQL to transfer data from database
into program and vice versa
Cannot be used to represent database objects, such
as table names or column names
Declared in a special Section of the program initially
Prefixed by colon : , when used in the program
Additional variables called indicators can be
associated with these variables to indicate the status of
data transfer like:
Indicating NULL values in Db
Truncation of data etc..
CS 53
Ramasesha Mudigere
169
Host Variables DECLARE Section
CS 53
Ramasesha Mudigere
170
Indicator variables
Integer Variables Used to indicate the status of data
transfer between host variables and the database
Can be associated with any host language variable
Used immediately following associated host variable
with a colon (:) separating two variables.
0 in Indicator variable indicates successful transfer
Non-zero value to indicate conditions like
CS 53
Data truncation (> 0)

Null value transfer ( < 0)
Type mismatch etc
Ramasesha Mudigere
171
Declaration & Use of Indicator variables

EXEC SQL BEGIN DECLARE SECTION;
char address[51];
short addressInd;
EXEC SQL END DECLARE SECTION;
addressInd = -1;
EXEC SQL UPDATE PrivateOwner
SET address = :address :addressInd
WHERE ownerNo = CO21;
CS 53
Ramasesha Mudigere
172
SQLCA Area
A data structure to be included in the program by the
programmer
Used ( initialized) by DBMS to indicate the status of
every SQL execution in the program
Set by DBMS after execution of each DBMS command
Contains several fields of which SQLCODE is of
special interest
CS 53
0 - statement executed successfully;

< 0 - an error occurred;
> 0 - statement executed successfully, but an exception
occurred, such as no more rows returned by SELECT.
Ramasesha Mudigere
173
CS 53
Ramasesha Mudigere
174
Embedding SQL Commands

Starts with identifier, usually EXEC SQL
Ends with terminator dependent on host language:
Ada, C, and PL/1: terminator is semicolon (;)
COBOL: terminator is END-EXEC
Fortran: ends when no more continuation lines.
Embedded SQL can appear anywhere an executable

host language statement can appear
CS 53
Ramasesha Mudigere
175
Embedded SQL Example
CS 53
loop = 1;
while (loop) {
prompt (Enter SSN: , ssn);
EXEC SQL
select FNAME, LNAME, ADDRESS, SALARY
into :fname, :lname, :address, :salary
from EMPLOYEE where SSN == :ssn;
if (SQLCODE == 0) printf(fname, );
else printf(SSN does not exist: , ssn);
prompt(More SSN? (1=yes, 0=no): , loop);
END-EXEC
}
Ramasesha Mudigere
176
Cursor
Used when multiple rows are Expected to be returned
by SQL query
Allows host language to access rows of query one at a
time
Acts as a pointer to a row of query result
Can be advanced by one to access next row
Using of cursor:
CS 53
Must be declared
opened before it can be used and
Must be closed to deactivate it after it is no longer required
Ramasesha Mudigere
177
Cursor Declaration & Opening

Declared using the SQL DECLARE as follows:
EXEC SQL DECLARE
propertyCursor CURSOR FOR
SELECT propertyNo, street, city
FROM PropertyForRent
WHERE staffNo = SL41;
OPEN statement opens specified cursor and positions

it before first row of query result:
EXEC SQL OPEN propertyCursor;
CS 53
Ramasesha Mudigere
178
Use of Cursor to access data

FETCH retrieves next row of query result table:
EXEC SQL FETCH propertyCursor
INTO :propertyNo, :street, :city
FETCH is usually placed in a loop.
When there are no more rows to be returned,
SQLCODE is set to NOT FOUND
Cursor needs to be closed at the end of use using
EXEC SQL CLOSE propertyCursor;
CS 53
Ramasesha Mudigere
179
Dynamic SQL
SQL that does not need to be hard coded into the program
Built with character string on the fly
Example of using Dynamic SQL in C
char * sqlprog = update account
setbalance = balance * 1.05
where account-number = ?
EXEC SQL preparedynprogfrom :sqlprog;
characcount [10] = A-101;
EXEC SQL execute dynprogusing :account;
The dynamic SQL program contains a ?, which is a place
holder for a value that is provided when the SQL program is
executed.
CS 53
Ramasesha Mudigere
180
4. SQL Programming Component

1. DECLARE
To define CURSORS to access database
2. OPEN
To enable / activate CURSORS
3. FETCH
To get data rows based on CURSOR
4. CLOSE
To disable /deactivate the cursor
CS 53
Ramasesha Mudigere
181
Dynamic Embedded SQL

With static embedded SQL, cannot use host variables
where database object names required.
Dynamic SQL allows this.
Idea is to place complete SQL statement in a host
variable, which is passed to DBMS to be executed.
If SQL statements do not involve multi-row queries,
use EXECUTE IMMEDIATE statement:
sprintf(buffer, UPDATE Staff
SET salary = salary + %f
WHERE staffNo = SL21 , increment);
EXEC SQL EXECUTE IMMEDIATE :buffer;
CS 53
Ramasesha Mudigere
182
PL / SQL The other alternative

A new language defined with programming constructs
Like:
Variable definition
Control branching
Looping
etc..
That allows for writing programs

Pioneered by ORACLE
Now part of SQL 1999 Standard (SQL /PSM)
CS 53
Ramasesha Mudigere
183
SQL/PSM: An Example
CREATE FUNCTION DEPT_SIZE (IN deptno INTEGER)
RETURNS VARCHAR[7]
DECLARE TOT_EMPS INTEGER;
SELECT COUNT (*) INTO TOT_EMPS
FROM SELECT EMPLOYEE WHERE DNO = deptno;
IF TOT_EMPS > 100 THEN RETURN HUGE
ELSEIF TOT_EMPS > 50 THEN RETURN LARGE
ELSEIF TOT_EMPS > 30 THEN RETURN MEDIUM
ELSE RETURN SMALL
ENDIF;
CS 53
Ramasesha Mudigere
184
Call Level Interfaces (CLI)

A more modern approach to the host language SQL
connection
C or other language program creates SQL statements
as character strings and passes them to functions that
are part of a library
Similar to what really happens in embedded dynamic
SQL implementations
A part of the SQL standard, that Provides easy access
to several databases within the same program
Certain libraries (e.g., sqlcli.h for C) have to be
installed and available
CS 53
Ramasesha Mudigere
185
Steps in SQL /CLI Programming

1. Load SQL/CLI libraries
2. Declare record handle variables for the above
components (called: SQLHSTMT, SQLHDBC,
SQLHENV, SQLHDEC)
3. Set up an environment record using SQLAllocHandle

4. Set up a connection record using SQLAllocHandle
5. Set up a statement record using SQLAllocHandle
CS 53
Ramasesha Mudigere
186
Steps in SQL /CLI Programming (Contd.)

6. Prepare a statement using SQL/CLI function
SQLPrepare
7. Bound parameters to program variables

8. Execute SQL statement via SQLExecute
9. Bound columns in a query to a C variable via
SQLBindCol
10. Use SQLFetch to retrieve column values into C

variables
CS 53
Ramasesha Mudigere
187
ODBC A CLI from Microsoft

An extended version of SQL/CLI originally designed for
windows, but got extended to other platforms later by
other vendors like Visigenic, Intersolv, OpenLink etc
Current version is 3.5 the unicode enabled version
Has over 80 calls that fall into 3 conformance levels
CS 53
Core that lets you connect to DB, execute SQL , fetch

results, Commit & rollback transactions, handle exceptions
and terminate connections
Level 1 - that lets you retrieve info from DB catalog, fetch
large objects & deal with driver specific functions
Level 2 that lets you retrieve data using cursors (forward
& backward)
Ramasesha Mudigere
188
ODBC (Contd.)
Very successful API, with virtually every database
vendor providing an ODBC driver for its database,
However
Each vendor have their own proprietary native API
(Oracle Oracle CLI, IBM ESQL/DRDA, Sybase
Sybase open client MS- SQL ODBC)
Vendors have added their own extensions to ODBC

driver making it non-standard again !!!
Specs are controlled by Microsoft ; not open standards
Future is uncertain with Microsoft pushing its OLE
based ADO as new standards
CS 53
Ramasesha Mudigere
189
JDBC A Java based CLI
A portable SQL CLI written entirely in JAVA

Provides 2 major sets of interfaces ( similar to ODBC)
Lets you write DBMS-independent Java code (with a
fair amount of effort) ; Part of Java Core
1. An application Interface that lets you access DBMS
services in an independent manner
2. A driver interface that DBMS vendors must adapt to
their particular databases
Uses a driver manager to automatically load the right
JDBC driver to talk to a given Database
All major DB vendors have agreed to support JDBC
CS 53
Ramasesha Mudigere
190
End of Chapter
-4-
Structured Query Language
CS 53
Ramasesha Mudigere
192
SQL 92 Features
SQL-92 -- Ratified in late 1992
SQL client / server connections (SET CONNECTION )

More granular transaction control
CS 53
Read only transactions

Multiple isolation levels ( Read-committed, read-uncommited etc)
Embedded SQL support for C, ADA etc..

Support for dynamic SQL
New data types ( BLOBS, VARCHAR, DATE, TIMESTAMP)
Support for temporary tables
Standardized error codes & Diagnostics & meta-data structures
Support for JOIN operators ( Outer JOIN, Inner JOIN etc..)
Domain checks and constraints
Miscellaneous improvements like backward & forward cursors, new
string functions
Ramasesha Mudigere
. Contd.
193
SQL 1999 Features

New Semantics:
New data types:

Large Objects (LOBs,
CLOBs, BLOBS)
BOOLEAN
Composite Types: ARRAY
and ROW
Distinct Types
New Predicates:
SIMILAR TO 'pattern'
(analogous to regular
expression matching)
DISTINCT objects
CS 53
Broader classes of views

Recursive queries
Locators
Enhanced Security:
Roles
Active Database:
Triggers and Stored
Procedures
Object Orientation:
Structured types & UDT

Methods, with Functional or
Dot Notations
Objects and REF Types
Ramasesha Mudigere
194
SQL 1999 Features
"
?
e
r
o
New Semantics:
New data types:
m
y
n classes of views
Broader
w
A
e
Large Objects (LOBs,
i
)
d tm queries
v
r
r
Recursive
a
e
h
CLOBs, BLOBS)
d
v
.
n Locators
o
1
a
0
t
n
y
a
S
h
BOOLEAN
l
s
6
a i0Enhanced
e Security:
1
e
s
v
i
r
R
/
g do
Composite Types: ARRAY
m
A
Roles
d
o
n
n
L
c
e
a
.
Q an esActive Database:
v
and ROW
l
S
a
s
d
r
r
I
t
e
.
u
"
t
,
v
Distinct Types
w
a
Triggers and Stored
e
e
w
s
f
e
l ://w
y
Procedures
9
c
i
b
9
t
New Predicates:
r
9 ce
p
t
a
1
t
Object Orientation:
e
n
h
&
(
h
a
i
2
T SIMILAR TO 'pattern'
l
9
p
Structured types & UDT
9
m
1
o
(analogous
to regular
L
c
Methods, with Functional or
Q
f
o
S
expression
matching)
Dot Notations
s
t
li s
DISTINCT objects
CS 53
Objects and REF Types
Ramasesha Mudigere
195
SQL Standards Criticisms

A former member of the Committee (Joe Celko) states:
"Seriously, the SQL-99 Standard is going nowhere
Another member (Michael Gorman) wrote a critical article

titled, "Is SQL A Real Standard Anymore?"
(http://www.tdan.com/i016hy01.htm) with following predictions:
"Competition among DBMS vendors will drop, prices will rise,
portability of programs, data & trained staff will end
The key criticism is that there is no enforcer to ensure that

vendors really follow the standard
Without enforcement, fission occurs as it did with Unix and
It's too easy for a vendor to reap benefit from the standard by
claiming compliance without really providing it.
CS 53
Ramasesha Mudigere
196
SQL Data types

Data Types
Binary
String
Character
String
UDT
Double-Byte
Char String
DateTime
Numeric
decimal
DATE time timestamp

integer
CHAR
BLOB
CS 53
CLOB
GRAPHIC
DBCLOB
VARCHAR
VARGRAPHIC
DECIMAL
floating
point
SMALLINT
INTEGER
Ramasesha Mudigere
REAL
DOUBLE
197
Relational Database Schema for Examples

1
2
3
4
5
CS 53
Ramasesha Mudigere
198
Populated Database
2
5
4
6
CS 53
Ramasesha Mudigere
199

04 SQL

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

04 SQL

Caricato da

Copyright:

Formati disponibili

-4The Structured Query Language

SQL-86 (A.k.a. SQL1)

issued by ANSI and ISO in 1986

SQL-89 - Minor revision like

SQL-92 ( A.k.a. SQL2) -- Ratified in late 1992

Defines three levels of compliance - Entry, Intermediate, &

The current version of SQL standards

SQL 2003 ( A.k.a. SQL4)

1. SQL DML Component

To retrieve rows from tables

To modify the rows of tables

To add new rows to tables

To remove rows from tables.

2. SQL DDL Component

3. SQL Admin Component

4. SQL Programming Component

5. SQL Transaction Component

SQL Statements A Summary

SQL Statements Coding Format

Tabs and Indentations will promote readability.

1. SQL DML Component

To retrieve rows from tables

To modify the rows of tables

To add new rows to tables

To remove rows from tables.

An SQL SELECT Query

< table list [alias] [, ...] >

< condition > ]

< grouping attribute(s) > ]

< group condition > ]

< attribute list> ]

Only the first two, SELECT and FROM, are mandatory

Clauses of SELECT Statement

Simple SELECT command

SELECT * FROM EMPLOYEE

Lists only unique DNO ( 5, 4 and 1)

SELECT ALL DNO FROM EMPLOYEE

Simple SELECT with computed fields

Simple SELECT with AS Clause

SELECT SSN, LNAME NAME, SALARY *12

SELECT Statement with WHERE Clause

SELECT Statement with compound Condition

The WHERE clause in SQL can include

SELECT Statement with compound

SELECT Statement with compound

Range Search Condition

Range Search Condition - Example

Pattern Matching Condition

Wildcards vary depending on DBMS

Pattern Matching Condition

Set Membership Condition

List the customer number, last name, and first name

Set Membership Condition

ORDER BY using multiple keys

List keys in the order of importance in the ORDER BY

ORDER BY with multiple keys

Handling NULL Values

signifies an unknown value or a value that does not exist.

Aggregate operators in SQL

returns sum of values in specified column.

returns average of values in specified column.

returns smallest value in specified column.

Returns the number of non-null occurrences of a column