Create table

Creation of Table
Presented By
Parteek Bhatia
Assistant Professor
Department of Computer Science & Engineering
Thapar University, Patiala

What is Table?
A table can represent a single entity (entity is an
object whose information is stored in database)
that you want to track within your system. Each
entity or table has number of characteristics. The
table must have a unique name through which it can
be referred to after its creation .The
characteristics of the table are called its
attributes. These attributes can hold data.

What is Table?
This type of a table could represent a list of the
employees within your organization, or the orders placed
for your company's products. A table has one attribute,
which identifies each record uniquely; this attribute is
called as Primary Key. Each value in the Primary Key
attribute is unique and it cannot be NULL. Each record of
the table is called as tuple.
A table can also represent a relationship between two
entities. This type of a table could portray the association
between employees and their job skills, or the relationship
of products to orders. Within the tables, foreign keys are
used to represent relationships.

Table creation rules
•Table name and Column name can be 1 to 30 characters
long. First character must be alphabetic, but name may
include letters, numbers and underscores.
•Names must contain only the characters A-Z, a-z, 0-9,
_(underscore),$ and # .
•Names must not be an Oracle Server reserved word.
•Names must not duplicate the names of other objects
owned by the same Oracle server user.
•Table name is not case sensitive.

SYNTAX
• Create Table [schema] tablename
(
{Column datatype [DEFAULT expr] [column constraint],
column datatype[DEFAULT expr] [column constraint],…
[table_constraint]}
);
Create Table table_name(
column_name datatype(width),
column_name datatype(width),…….);

Schema
It is the schema to contain the table. If it is omitted,
Oracle creates the table in creator’s own schema.
Table name
It is the name of the table to be created. Table name
and Column name can be 1 to 30 characters long. First
character must be alphabetic, but name may include
letters, numbers and underscores. The table name
cannot be the name of another object owned by same
user and cannot be a reserved word.
Column
Specifies the name of a column of the table. The number
of columns in a table can range from 1 to 1000.

Datatype
It is the datatype of a column. Specifies the type
and width of data to be stored in the column.
Default
It specifies a value to be assigned to the column if
a subsequent INSERT statement omits a value for
the column. The datatype of the expression must
match the datatype of the column. A DEFAULT
expression cannot contain references to other
columns, the pseudo columns CURRVAL,
NEXTVAL, LEVEL and ROWNUM or data
constants that are not fully specified.

SQL> Create table student
(Name varchar2 (20),
Class varchar2 (15),
Roll_no number (4),
Address varchar2 (30));

Role of Constraints to achieve data
integrity
To understand the role of constraints in database let us
consider a case, when we appear for some interview,
there are certain constraints for our qualification.
Person who satisfies the given qualification constraints,
only those person are eligible for interview, so these
restrictions are called constraints which are to be
enforced to select the correct candidate. Such
limitations have to be enforced on the data to achieve
the integrity (Correctness). The data, which does not,
satisfies the conditions will not be stored, this ensures
that the data stored in the database is valid and has
gone through the integrity rules which prohibit the
wrong data to be entered into the database.

Categories of Constraints
There are two types of constraints, these are:
• Column constraints
• Table constraints.
Column Constraints
When we define constraints along the definition
of the column while creating or altering the table
structure, they are called as column constraints. Column
constraints are applied only to the individual columns.
These constraints are applied to the current column .A
column level constraint cannot be applied if the data
constraint spans across multiple columns in a table.
For example: Empno of emp table is a primary key, which
is column level constraint.

Table Level Constraints
Table level constraints are applied to more than
one column of a table. These constraints are
applied after defining all the table columns when
creating or altering the structure of the table.
A table level constraint can be applied if the
data constraint spans across multiple columns in
a table.

Table Level Constraints
• For example: In case of bank database the
account number field of account holder table is
not sufficient to uniquely identify the entities
because two person can have joint account and
hence repeating the value of account number
once for each account holder, so in this case
we have to apply primary key constraint on
combination of account number field and name
field to achieve the uniqueness. Now in this
case the primary key constraint is applied on
more than one column of the table so this is the
table level constraint.

Types of Constraints
NULL/NOT NULL
Specifies if the column must contain value or might not contain any.
By default all columns in a table allow nulls, i.e. absence of a value
in a column. NOT NULL specifies that all rows in the table to have
value for specified column. All NOT NULL columns are mandatory
fields. Row of data will not be accepted for the table unless columns
so defined have data in them. NULL columns might not contain any
data and can be left empty. NULL should not be however specified
for numeric fields as any arithmetic with NULL results in a NULL
value.
Syntax:
Columnname datatype (size) [constraint
constraintname] NOT NULL

Example
• Create a table student with the fields name, roll_no,
address and phone number.
SQl> Create table student
(Name varchar2 (20) CONSTRAINT NN_NAME NOT
NULL,
Roll_no number (5) NOT NULL,
Address varchar2 (40),
phone_no varchar2 (10));

Example
SQL>CREATE TABLE sale(
code char(4) NOT NULL,
description VARCHAR2(15) CONSTRAINT desc_item NOT
NULL,
quantity NUMBER(5), sale_date DATE, notes CHAR(15) NULL);
Above statement creates a table with fields code and description
which is mandatory i.e. cannot be left empty and notes which is
optional i.e. might not contain data.

Unique constraint
The unique key allows unique values to be entered into
the column i.e. every value is a distinct value in that
column. When we apply a unique constraint on a group of
columns then each value of the group should be unique,
they must not be repeated. A table can have more than
one unique key. This constraint can be applied both at
the table level and the column level. The syntax is:
Column level syntax:
constraintname] UNIQUE

Unique constraint
Example
• Create a table student with roll_no not null, unique
name, unique address and unique phone number.
(Roll_no number (5) NOT NULL,
name varchar2(20) contraint un_name unique,
Address varchar2 (40) unique,
phone_no varchar2 (10) unique);
Table level syntax of unique constraint
[Constraint constraintname] Unique (columnname [,
columnname, …….])

Unique constraint
• Create a table student with roll_no not
null and the combination of name, address and
phone number must be unique.
(Roll_no number (5) NOT NULL,
Name varchar2 (20),
Address varchar2 (40),
phone_no varchar2 (10),
Constraint tb_un UNIQUE (name, address,
phone_no));

Example
SQL>CREATE TABLE item(
code CHAR(4), description CHAR(10), rate NUMBER,
UNIQUE (CODE, DESC));
The above statement creates a table item in which fields
code and description cannot be duplicate or same for more
than one record. In this example unique has been
specified as table constraint.

Primary Key Constraint
A primary key is used to identify each row of the table
uniquely. A primary key may be either consists of a single
field or group of fields. It is a combination of both the
unique key constraint as well as the not null constraint i.e.
no column with the primary key constraint can be left blank
and it must contain unique value in each row. We can
declare the Primary key of a table with NOT NULL and
UNIQUE constraint .SQL supports primary keys directly
with the Primary Key constraint. When primary key is
applied on a single field it is called as Simple Key and when
applied on multiple columns it is called as Composite
Primary Key. In a table there is only one primary key.

Columnname datatype(size) [ constraint_name]
PRIMARY KEY
• Create a table student with roll_no as the Primary
Key, unique name, address cannot be NULL and phone
number.
(Roll_no number (5) Primary key,
Name varchar2 (20) Unique,
Address varchar2 (40) Not Null,
phone_no varchar2 (10));

Table level syntax of Primary Key constraint
PRIMARY KEY (columnname [, columnname , ……])
• Create a table student with name and class
as the composite primary key, address and
phone_no are the other fields.
(Name varchar2 (20),
Class varchar2 (15),
Address varchar2 (40) Not Null,
phone_no varchar2 (10)
PRIMARY KEY (name, class));

Examples
SQL>CREATE TABLE student (
name CHAR(10) PRIMARY KEY, class NUMBER, marks
NUMBER);
SQL>CREATE TABLE emp(
empno NUMBER(4) PRIMARY KEY,
ename VARCHAR2(10), job VARCHAR2(9), mgr NUMBER(4),
hiredate DATE,
sal NUMBER(7,2), comm NUMBER(7,2), deptno NUMBER(2));

Check Constraint
Check constraints allow Oracle to verify the validity of
data being entered on a table against a set of
constants. These constants act as valid values. The
Check constraint consists of the keyword CHECK
followed by parenthesized conditions. Check constraints
must be specified as a logical expression that evaluates
either to TRUE or FALSE. If an SQL statement causes
the condition to be false an error message will be
displayed.
constraintname] CHECK (logical expression)

Check constraint
• Create a table EMP1 with empid as primary
key, phone number as unique attribute and the
salary of all the employees must be greater than
5000 with name and class as the composite
primary key ,address and phone_no are the other
fields.
SQL> CREATE TABLE emp1
(empid number (10) PRIMARY KEY,
salary number (10, 3) CHECK (salary > 5000),
home_phone number (12),
CONSTRAINT uk_phone UNIQUE
(home_phone));

If, for example, someone tries to create an employee
row for the table defined earlier with a salary of Rs.
1000, Oracle will return an error message saying that
the record data defined for the SALARY column has
violated the check constraint for that column.

Check constraint
• Create a table emp1 with empid as primary
key, phone number as unique attribute and the
department of the employees must be either
general or accounts, address and phone_no are
the other fields.
SQL> CREATE TABLE emp1
(empid number (10) PRIMARY KEY,
deptname varchar2 (20) CHECK
(deptname in (‘general’,’accounts’)),
home_phone number (12) UNIQUE);

Limitations of Check Constraints:
Check constraints have a number of limitations,
these are:
• A column level check constraint cannot refer
to another column of any table.
• It cannot refer to special keywords that can
have values in them, such as user, sysdate, or
rowid.

• Example 1
CREATE TABLE emp1
(empid number (10),
salary number (10, 3),
Comm. number (10, 3),
home_phone number (12),
CONSTRAINT pk_empid
PRIMARY KEY (empid),
CONSTRAINT uk_phone UNIQUE (home_phone),
CHECK (salary > comm));

• Example 2
CREATE TABLE emp1
(empid number (10),
salary number (10, 3),
Home_phone number (12),
CONSTRAINT pk_empid
PRIMARY KEY (empid),
CONSTRAINT uk_phone UNIQUE (home_phone),
CHECK (salary < 500000));

Example
CREATE TABLE emp(
code CHAR(4) PRIMARY KEY, name VARCHAR2(15) NOT
NULL,
department CHAR(10) CHECK (department IN
(‘mkt’,’sales’,’Acct’))
age NUMBER CHECK (age between 18 and 55),
basic NUMBER);
Using in clause with a column the valid values to be entered in a
column can be specified. Using between clauses the valid range can
be specified.

Default constraint
The default value constraint allows the user to
insert the values in the columns where the user
do not want to insert the value .The most
common example of this constraint is NULL
value, which is inserted in columns not defined
with the NOT NULL constraint. This is not
actually a constraint, it only specifies that a
value should be inserted in a column if the user
does not enter a value .The default values
assignments are defined at the time of create
table command. The datatype of the default
value should match the datatype of the column.

The syntax is:
Columnname datatype (size) DEFAULT value;
(roll_no number(5) primary key,
Name varchar2 (20) not null,
Marks number (3) default 100,
address varchar2 (30) not null));

Foreign Key constraint
A foreign key is a kind of constraint, which
establishes a relationship among tables. A
foreign key may be a single column or the
combinations of columns, which derive their
values, based on the primary key or unique key
values from another table. A foreign key
constraint is also known as the referential
integrity constraint, as its values correspond to
the actual values of the primary key of other
table.

A table in which there is a foreign key, it is
called as the detail table or the child table and
the table from which it refers the values of
the primary key, it is called as the master table
or the parent table .A foreign key must have
the corresponding primary key or unique key
value in the master table. Whenever you insert
a value in the foreign key column, it checks the
value from the primary key of the parent table.
If the value is not present, it gives an error.
The parent table can be referenced in the
foreign key by using the references option.

Whenever a user tries to delete a record in
the master table, which is being referenced in
the child table, or the record exists in the
child table, then the Oracle displays an error
message. To avoid this error message, the ON
DELETE CASCADE option is used which will
delete the record in the detail table when the
user deletes the record in the master table.
The syntax of the foreign key at the column
level is:
Columnname datatype (size) references
tablename [(columnname)] [ON DELETE
CASCADE]

Examples
Referenced table
SQL>CREATE TABLE item( item_code CHAR(4) PRIMARY
KEY,
description VARCHAR(20,
rate NUMBER(6,2));
Dependent table
SQL>CREATE TABLE sales(
invoice_no NUMBER(5), item_code CHAR(4) REFERENCES
item(item_code),
qty_sold NUMBER(5));
Item_code can be inserted in sales table only if it has references in
item table i.e. it exists in the item table.

Referenced table
SQL>CREATE TABLE dept (DEPTNO NUMBER(2)
PRIMARY KEY,
DNAME VARCHAR2(14), LOC VARCHAR2(13))
Dependent table
SQL>CREATE TABLE emp(
empno NUMBER(4) PRIMARY KEY,
ename VARCHAR2(10), job VARCHAR2(9), mgr
NUMBER(4), hiredate DATE,
sal NUMBER(7,2), comm NUMBER(7,2), deptno
NUMBER(2) REFERENCES dept(deptno) on delete
set null);

SQL> Create table emp_detail
(Empno number (4) primary key,
Ename varchar2 (20) not null,
Hiredate date not null,
Deptno number (2),
dname varchar2 (10),
Salary number (6, 2)
Foreign key (deptno,dname) references
dept(deptno,dname)) ;
So in this table deptno and dname is the foreign key,
which references the corresponding fields in the dept
table.

Other Options with foreign key

The user_constraints table
A user can define multiple number of constraints
in a table and if the user wishes to see the name
of the constraints ,they are available in the
USER_CONSTRINTS table .The DESCRIBE
option only gives the name of the fields in the
table and NOT NULL constraint associated with
the column . So the user_constraints table
provides information pertaining to the names of
all the constraints on the table .The following
fields is there in the user_constraints table:

So you can see the name of the constraints associated with
the table by writing a simple SQL query.
SQL>Select owner, constraint_name, r_owner,
r_constraint_name from user_constraints where
table_name = ‘EMP_DETAIL’;
The Output is

SQL>Select * from user_constraints where
table_name = ‘EMP’;

USER_CONS_COLUMNS
• It can be used to see the list of constraints with
the column name on a table.
SELECT * FROM USER_CONS_COLUMNS;

Creating a Table with rows from another table
A table is created using CREATE TABLE statement with
rows in place, derived from another table.
Syntax
CREATE TABLE table_name
[(Column name……………)]
AS SELECT statement;

The table will be created with the specified columns and
the rows retrieved by the SELECT statement inserted into
it. If all the columns in the SELECT statement have well
defined names, (that is, no expressions, and so on), the
column specifications may be omitted. If column
specifications are given, the number of columns must
equal the number of items in the SELECT list. Constraints
information is inherited from the selected table. Data type
for the column cannot be specified.

Example
• Create a table emp2 from emp table having employee number,
name, mgr, and sal of employees in department 10 and 20.
SQL>CREATE TABLE emp2
AS
SELECT empno, ename, mgr, sal FROM emp
WHERE deptno in (10,20);
Table created
The above table contains all the records present in emp table having
deptno 10 or deptno 20. Also the constraints attached to columns in
emp table are attached to the columns of emp2 table.

• Create a table Salary from table emp having empno,
ename and salary details.
SQL> CREATE TABLE salary(employee_number,Name,
Salary)
AS
SELECT empno, ename, sal FROM emp;
Table created
The above table contains records of table emp having
empno, ename, salary details.

Creating table from an existing
Table
In addition to creating table as above, we can
also create a new table from the existing table
also .We apply the AS sub-query clause to both
create the table and insert rows returned
from the sub query. For example;
SQL> Create table emp1
AS
Select empno, ename, hiredate, sal from EMP
where comm IS NULL;

References
Simplified Approach to Oracle
Kalyani Publishers
By
Author Parteek Bhatia

Create table

More Related Content

What's hot (20)

Similar to Create table (20)

More from Nitesh Singh (20)

Recently uploaded (20)

Create table