Introduction to Relational Databases

Informal look to Relational Databases

The relational model is having the three aspects:

� Structural aspect:

� The data in the database is perceived by the user as

tables

� Integrity aspect:� Integrity aspect:

� Those tables satisfy certain integrity constraints

� Manipulative aspect

� The operators are available to the user for manipulating those tables

A formal definition� Relational databases store data in the form of tables

� The rows of a table are called as tuples� The columns of a table are known as attributes� Every attribute has a data type associated with it� Every attribute has a domain which provides the set of possible

values� Tables are called as relations� The table names are called as relational variables (relvar)

A Formal Definition

Domains

� A relation may be regarded as a set of tuples

� A domain has a logical definitions: eg., “phone_number” are the set of 10 digit numbers

� A domain defines posible values of an attribute, may have a data-type or a format defined for itdata-type or a format defined for it

� An attribute is the name of a role played by some domain. Eg.,domain Date may be used to define attributes “Invoice-date” and “Payment-date”

Relation schema

� A relation schema is used to describe a relation� A relation schema R, denoted by R(A1,A2,...,An):

� where R is a relation name and� A1,A2,...,An - a list of attributes

� The degree of a relation is the number of attributes� Instance r of R : a specific “value” or population of R� Instance r of R : a specific “value” or population of R

Characteristics� Ordering of tuples in a relation:

� The tuples are not considered to be ordered, even though they appear to be in the tabular form

� Ordering of attributes:� We will consider the attributes in R(A1,A2,...,An) and the� We will consider the attributes in R(A1,A2,...,An) and the

values in t = <v1,v2,...,vn> to be ordered.

� Values in a tuple:� All values are considered atomic (indivisible).

� A special null is used to represent unknown or inapplicable values

Characteristics

Relational Integrity Constraints

� Constraints are conditions that must hold on all valid relation instances.

� Key Constraints

� Super Key

� Key� Key

� Candidate key

� Entity integrity Constraints

� Primary key

� Referential integrity Constraints

� Foreign key

� Another implicit constraint is the domain constraint Every value in a tuple must be from the domain of its attribute (or it could be null, if allowed for that attribute)

SuperKey

� Superkey of R:

� Is a set of attributes SK of R with the following condition:

� No two tuples in any valid relation state r(R) will have the same value for SK

� That is, for any distinct tuples t1 and t2 in r(R), t1[SK] ≠ � That is, for any distinct tuples t1 and t2 in r(R), t1[SK] ≠ t2[SK]

� This condition must hold in any valid state r(R).

� Every relation has at least one default superkey—the set of all its attributes.

� {Ssn, Name, Age}—is a superkey

Key

� Key: A set of attributes K, whose values uniquely identify a tuple in any instance. And none of the proper subsets of K has this property.

� {ssn} is a key for student relation.

� Can be called as minimal superkey, that is, a superkey from � Can be called as minimal superkey, that is, a superkey from which we cannot remove any attributes and still have the uniqueness constraint can hold.

� The superkey {Ssn, Name, Age}is not a key of STUDENT because removing Name or Age or both from the set still leaves us with a superkey.

� Any superkey formed from a single attribute is called as key.

Candidate Keys� A relation can have more than one key.� Each of the keys is called a candidate key

� Consider a car relation where has two candidate keys: License_number and Engine_serial_number.

Primary Key

� A given relation may have two or more candidate keys

� Exactly one of those keys be chosen as the primary key, the others are then called alternate keys

� The CAR relation with two candidate keys: LicenseNumber and EngineSerialNumberEngineSerialNumber

Primary Key

� We could choose either {ID} or

{First, Last} as the Primary Key.

� ID is more convenient as it is a

single column and we know it will

always be unique (what happens

if another John Smith is added?)

Entity Integrity

� The primary key attributes PK of each relation schema R cannot have null values in any tuple of r(R)

� This is because primary key values are used to identify the individual tuples

� Other attributes of R may be constrained to disallow null, even � Other attributes of R may be constrained to disallow null, even though they are not members of the PK

Foreign keys

� Foreign key is a set of attributes of some relvar R2 whose values are required to match values of some candidate key of some relvar R1.

� Can be specified using two terms

� the referencing relation � the referencing relation

� the referenced relation

� A foreign key is a set of attributes, say FK, of referencing relation whose values are required to match values of candidate key CK of referenced relation.

� The FK and CK are the same data type.

� The FK must appear as a value of CK.

� FK is simple or composite according as CK .

Referential Integrity

� Referential integrity – the database must not contain any unmatched foreign key values.

� The value in the foreign key column FK of the referencing relation can be either:

� a value of an existing primary key PK in the referenced � a value of an existing primary key PK in the referenced relation, or

� a null

� In other words, the constraint says simply:

� If B references A, then A must exist

Referential Integrity

{DID} is a Candidate Key for {DID} is a Foreign Key in Employee

Department Employee

{DID} is a Candidate Key for Department – Each entry has a unique value for DID

{DID} is a Foreign Key in Employee- each Employee’s DID value iseither NULL, or matches an entry inthe Department relation.

FOREIGN KEY <item commalist> REFERENCES <relvar name>

Referential Integrity

{ID} is a Candidate Key for Employee,

and {Manager} is a Foreign Key, which

refers to the same relation – every

tuple’s Manager value is either NULL or

matches an ID value

FOREIGN KEY Manager REFERENCES Employee(ID)

Referential Integrity

� When relations are updated, referential integrity can be violated� When a referenced tuple is updated / deleted, an exception will

be raised� To compensate this, the options are:

� Cascade: Actions cascaded to the matching tuples� Cascade: Actions cascaded to the matching tuples� Restrict: Restricted to the case where there are no matching

tuples� Nullify: Make values NULL� Triggers: Actions defined by the user

Referential Integrity

� What happens if

Marketing’s DID is

changed to 16 in

Department?

�� The entry for Accounts

is deleted from

Department?

Referential Actions – RESTRICT

� RESTRICT stops any action that violates integrity

� You cannot update or delete

Marketing or Accounts

� You can change Personnel

Department

� You can change Personnel

as it is not referenced Employee

Referential Actions – CASCADE

� CASCADE allows the changes made to flow through

� If Marketing’s DID is changed to 16 in Department, then the DIDs for John Smith and Mark Jones also change

Department

also change� If Accounts is deleted then so is

Mary BrownEmployee

Referential Actions – NULLIFY

� NULLIFY sets referenced values to NULL

� If Marketing’s DID changes then John Smith’s and Mark Jones’ DIDs are set to NULL

Department

NULL

� If Accounts is deleted, Mary Brown’s DID becomes NULL

Employee

SQL Facilities

� An SQL candidate key definition takes one of the following:� PRIMARY KEY ( <column name commalist> )� UNIQUE ( <column name commalist> )

� An SQL foreign key definition takes the form:� FOREIGN KEY ( <column name commalist> )� FOREIGN KEY ( <column name commalist> )REFERENCES <base table name> [ (<column name commalist>) ][ ON DELETE <referential action> ][ ON UPDATE <referential action> ]

� Thank you