Relational D Bs3

CIS3465: Professor Kirs Slide Number: 1

Chapter 2:The Relational Database Model

(With Considerable Modifications)

RDBMS Basics

CIS3465: Professor Kirs Slide Number: 2

RDBMS Basics

We already know that a Relational DataBase Management System (RDBMS) is:

• A DBMS Approach which manages data (logically) as a collection of tables where data, and data relationships, are represented by common values in related tables

• A Table (file) consists of records, which consist of attributes (fields), which relies on the storage of various data types

CIS3465: Professor Kirs Slide Number: 3

Consider some information the University maintains: Name Address SSN

Major Courses Taken Grades Received

Tuition Paid Tuition Owed Grants/Scholarships

HOW is this information stored?

You are an entity with attributes which vary. Within the University, different areas have different interests in you (i.e., the Registrar, the Bursar, etc.). Nonetheless, you are still part of the University as a whole.

HOW does this relate to a database?

RDBMS Basics

CIS3465: Professor Kirs Slide Number: 4

You are an entity Recordwith attributes Fieldswhich vary Fields can contain characters,

numbers, symbols, etc.Within the University, different areas, have different interests in you

Files

(i.e,. The Registrar, Bursar, etc.)

Nonetheless, you are still part of the University Database

RDBMS Basics

Database Components:

CIS3465: Professor Kirs Slide Number: 5

Database Components:

A Database consists of

Files, which contain

Records, which contain

Fields, which may consist of a variety of data types

Jones, Mary 234567890 102 3.87

Hernandez, Juan 123456789 72 2.42

• • •

• • •

• • •

• • •

• • • • •

• • • • •

• • • • •

• • • • •

• •

• •

• •

• •

• • •

• • •

• • •

• • •

• •

• •

• •

• •

• • •

• • •

• • •

• • •

• • •

• • •

• • •

• • •

• •

• •

• •

• •

• • • •

• • • •

• • • •

• • • •

Notice that there should always be a Key (Unique) Field

RDBMS Basics

CIS3465: Professor Kirs Slide Number: 6

RDBMS Basics

RDBMS Restrictions/Conventions: Each Relation MUST have a unique name

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

234567890 Chung, Mei 37 5th St. INFOSYS

345678901 Adams, John 54B Hague Accounting

456789012 Elam, Mary 123-22 E St. Accounting

•••••• •••••• •••••• ••••••

Table Student

Student Owed Department

103456678 1,502.36 Marketing

123456789 COBA219 Finance

456789012 COBA232 Accounting

•••••• •••••• ••••••

Table BalanceSID •••••• Depart

987654321 •••••• Finance

876543210 •••••• INFOSYS

765432109 •••••• Accounting

•••••• •••••• ••••••

Table Student

NOT Allowed

CIS3465: Professor Kirs Slide Number: 7

RDBMS Basics

RDBMS Restrictions /Conventions : Each Relation MUST have a unique name

Table Student

All Columns (Tuples) MUST have Unique names

StudentID Name Address MajorAddress

123456789 Saenz, Lupe 123 Mesa Arlington Finance

234567890 Chung, Mei 37 5th St. INFOSYSNew York

345678901 Adams, John 54B Hague Dallas Accounting

456789012 Elam, Mary 123-22 E St. Ft. Worth INFOSYS

One of the names MUST be changed

* NOTE: The same field names CAN be used in different Relations

CIS3465: Professor Kirs Slide Number: 8

RDBMS Basics

RDBMS Restrictions /Conventions : Each Relation MUST have a unique name

Table Student

All Columns (Tuples) MUST have Unique names

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa 24.34

234567890 Chung, Mei 37 5th St. INFOSYS

345678901 Adams, John 54B Hague Accounting

456789012 Elam, Mary 123-22 E St. INFOSYS

Unless this is stored as Character String “24.34” (and NOT as the real number 24.34), it MUST be changed

All Column Elements MUST be of the same data type

CIS3465: Professor Kirs Slide Number: 9

RDBMS Basics

RDBMS Restrictions /Conventions : Each Relation MUST have a unique name

Table Student

All Columns (Tuples) MUST have Unique names

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

All Column Elements MUST be of the same data type• Notice that this means each record requires the

SAME number of Bytes of Storage

A Character String Requiring 9 Bytes of StorageA Character String Requiring 20 Bytes of StorageA Character String Requiring 25 Bytes of StorageA Character String Requiring 10 Bytes of Storage

A Total 64 Bytes

of Storage

CIS3465: Professor Kirs Slide Number: 10

RDBMS Basics

RDBMS Restrictions /Conventions : Each Relation MUST have a unique name All Columns (Tuples) MUST have Unique names

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

234567890 Chung, Mei 37 5th St. INFOSYS

345678901 Adams, John 54B Hague Accounting

All Column Elements MUST be of the same data type The order of Rows is NOT important

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finace

345678901 Adams, John 54B Hague Accounting

234567890 Chung, Mei 37 5th St. INFOSYS

Is the Same as

CIS3465: Professor Kirs Slide Number: 11

RDBMS Basics

RDBMS Restrictions /Conventions : Each Relation MUST have a unique name All Columns (Tuples) MUST have Unique names

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

234567890 Chung, Mei 37 5th St. INFOSYS

345678901 Adams, John 54B Hague Accounting

All Column Elements MUST be of the same data type The order of Rows is NOT important

Each contains 64 Bytes

The Number of Bytes/Record Must be the same

CIS3465: Professor Kirs Slide Number: 12

RDBMS Basics

RDBMS Keys: Purpose

Define entity relationships Determination

Knowing the value of a key field means you also know (Determine) the values of the other fields

E.g., knowing StudentID means you know StudentName, StudentAddress, etc.

StudentID StudentName, StudentAddress(StudentID Determines StudentName and StudentAddress)

CIS3465: Professor Kirs Slide Number: 13

RDBMS Basics

RDBMS Keys: Purpose

Define entity relationships Determination

An attribute is functionally dependent on another if can be determined by that attribute

Functional Dependence

StudentIDStudentAddress

(StudentID Determines Student Address)

NOTE:

StudentID StudentAddress

(Two Students MAY live at the same address)

CIS3465: Professor Kirs Slide Number: 14

RDBMS Basics

RDBMS Keys: Each Relation MUST have a unique identifier or

PRIMARY KEYPRIMARY KEY

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

234567890 Chung, Mei 37 5th St. INFOSYS

345678901 Adams, John 54B Hague Accounting

456789012 Elam, Mary 123-22 E St. INFOSYS

•••••• •••••• •••••• ••••••

Table Student

No two students can have the same StudentID

CIS3465: Professor Kirs Slide Number: 15

RDBMS Basics

RDBMS Keys: Each Relation MUST have a unique identifier or

PRIMARY KEYPRIMARY KEY A COMPOSITE KEYCOMPOSITE KEY is a combination of keys (Multi-

key attributes) used to produce uniqueness

StudentName,StudentAddress StudentMajor(StudentName AND StudentAddress Determines StudentMajor)

NOTE: If Student Major is functionally dependent upon StudentName AND StudentAddress , BUT not on either StudentName or StudentName it is FULLY FUNCTIONAL DEPENDENT on the Concatenated key

(Attributes are fully functionally dependent on PRIMARY KEYS)

CIS3465: Professor Kirs Slide Number: 16

RDBMS Basics

RDBMS Keys: Each Relation MUST have a unique identifier or

PRIMARY KEYPRIMARY KEY A is COMPOSITE KEYCOMPOSITE KEY is a combination of keys

(Multi-key attributes) used to produce uniqueness A is SUPER KEYSUPER KEY is either a PRIMARY or

COMPOSITE KEY that uniquely identifies an entity

StudentID StudentAddressAND

StudentName,StudentAddress StudentMajor

Are BOTH Superkeys

CIS3465: Professor Kirs Slide Number: 17

RDBMS Basics

RDBMS Keys: Each Relation MUST have a unique identifier or

PRIMARY KEYPRIMARY KEY A is COMPOSITE KEYCOMPOSITE KEY is a combination of keys

(Multi-key attributes) used to produce uniqueness A is SUPER KEYSUPER KEY is either a PRIMARY or

COMPOSITE KEY that uniquely identifies an entity A is CANDIDATE KEYCANDIDATE KEY is any key or group of keys

that could become a SUPER KEY• StudentID is a candidate key• StudentName,StudentAddress is a candidate key• StudentID,StudentAddress is NOT a candidate key

(StudentID by itself is a CANDIDATE KEY)

CIS3465: Professor Kirs Slide Number: 18

RDBMS Basics

RDBMS Keys: Each Relation MUST have a unique identifier or

PRIMARY KEYPRIMARY KEY A is COMPOSITE KEYCOMPOSITE KEY is a combination of keys

(Multi-key attributes) used to produce uniqueness A is SUPER KEYSUPER KEY is either a PRIMARY or

COMPOSITE KEY that uniquely identifies an entity A is CANDIDATE KEYCANDIDATE KEY is any key or group of keys

that could become a SUPER KEY

A is SECONDARY KEYSECONDARY KEY is any field, or combination of fields, which does NOT yield a unique value• Used for retrieval/Narrowing purposes only • StudentName,StudentZip may yield several records

CIS3465: Professor Kirs Slide Number: 19

RDBMS Basics

RDBMS Keys: Each Relation MUST have a unique identifier or

PRIMARY KEYPRIMARY KEY A is COMPOSITE KEYCOMPOSITE KEY is a combination of keys

(Multi-key attributes) used to produce uniqueness A is SUPER KEYSUPER KEY is either a PRIMARY or

COMPOSITE KEY that uniquely identifies an entity A is CANDIDATE KEYCANDIDATE KEY is any key or group of keys

that could become a SUPER KEY

A is SECONDARY KEYSECONDARY KEY is any field, or combination of fields, which does NOT yield a unique value• Used for retrieval/Narrowing purposes only • StudentName,StudentZip may yield several records

CIS3465: Professor Kirs Slide Number: 20

RDBMS Basics

RDBMS Keys: In Order to relate two (or more) tables FOREIGN FOREIGN

KEYS KEYS must be used

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

234567890 Chung, Mei 37 5th St. INFOSYS

345678901 Adams, John 54B Hague Accounting

Table Student

Table DepartmentDeptName Chairman Telephone

INFOSYS Sircar, Sumit 555-1234

Marketing McDaniel, C. 555-2345

Accounting Courtney, H. 555-3456

A FOREIGN KEY in one tableis a PRIMARY/SUPER KEY In another

CIS3465: Professor Kirs Slide Number: 21

RDBMS Basics

Database Integrity: Maintaining wholeness and Unity Entity Integrity:

All Entries MUST be Unique No NULL values in primary key fields

StudentID Name Address Major

123456789 Saenz, Lupe 123 Mesa Finance

234567890 Chung, Mei 37 5th St. INFOSYS

123456789 Adams, John 54B Hague Accounting

456789012 Elam, Mary 123-22 E St. INFOSYS

Bush, G.W. 555 Austin Marketing

Table Student

Illegal Entries

CIS3465: Professor Kirs Slide Number: 22

RDBMS Basics

Database Integrity: Maintaining wholeness and Unity Entity Integrity Relational Integrity:

Foreign Key MUST have a valid entry in the corresponding table (or be NULL)

StudentID Major

123456789 Scamming

234567890 INFOSYS

345678901

Table Student Table Department

DeptName Chairman Telephone

INFOSYS Sircar, Sumit 555-1234

Marketing McDaniel, C. 555-2345

Accounting Courtney, H. 555-3456

???

NOT Allowed

CIS3465: Professor Kirs Slide Number: 23

RDBMS Basics

Database Integrity: Maintaining wholeness and Unity Entity Integrity Relational Integrity:

Foreign Key MUST have a valid entry in the corresponding table (or be NULL)

Table Student Table DepartmentDeptName Chairman Telephone

INFOSYS Sircar, Sumit 555-1234

Marketing McDaniel, C. 555-2345

StudentID Major

123456789 Accounting

234567890 INFOSYS

345678901 Finance

Primary Key entry CAN NOT be deleted if a foreign key refers to it

INFOSYS can NOT be deleted

CIS3465: Professor Kirs Slide Number: 24

RDBMS Basics

Basic RDBMS Operators: SELECT

Choose All or Specific values from a table

SELECT ALLYields

Table Student

StudentID Name Address

123456789 Saenz, Lupe 123 Mesa

234567890 Chung, Mei 37 5th St.

345678901 Adams, John 54B Hague

456789012 Elam, Mary 123-22 E St.

567890123 Bush, G.W 555 Austin

678901234 Clinton, B. 89 Congress

789012345 Bush, J. 555 Austin

Given:

StudentID Name Address

123456789 Saenz, Lupe 123 Mesa

234567890 Chung, Mei 37 5th St.

345678901 Adams, John 54B Hague

456789012 Elam, Mary 123-22 E St.

567890123 Bush, G.W 555 Austin

678901234 Clinton, B. 89 Congress

789012345 Bush, J. 555 Austin

CIS3465: Professor Kirs Slide Number: 25

RDBMS Basics

Basic RDBMS Operators: SELECT

Choose All or Specific values from a table

SELECT WHERE ADDRESS = “555 Austin”Yields

Table Student

StudentID Name Address

123456789 Saenz, Lupe 123 Mesa

234567890 Chung, Mei 37 5th St.

345678901 Adams, John 54B Hague

456789012 Elam, Mary 123-22 E St.

567890123 Bush, G.W 555 Austin

678901234 Clinton, B. 89 Congress

789012345 Bush, J. 555 Austin

Given:

StudentID Name Address

567890123 Bush, G.W 555 Austin

789012345 Bush, J. Austin

CIS3465: Professor Kirs Slide Number: 26

RDBMS Basics

Basic RDBMS Operators: PROJECT

Produce list of all values for a selected attribute

PROJECT STUDENTIDYields

Table Student

StudentID Name Address

123456789 Saenz, Lupe 123 Mesa

234567890 Chung, Mei 37 5th St.

345678901 Adams, John 54B Hague

456789012 Elam, Mary 123-22 E St.

567890123 Bush, G.W 555 Austin

678901234 Clinton, B. 89 Congress

789012345 Bush, J. 555 Austin

Given:

StudentID

123456789

234567890

345678901

456789012

567890123

678901234

789012345

CIS3465: Professor Kirs Slide Number: 27

RDBMS Basics

Basic RDBMS Operators: PROJECT

Produce list of of all values for a selected attribute

PROJECT STUDENTID, NAMEYields

Table Student

StudentID Name Address

123456789 Saenz, Lupe 123 Mesa

234567890 Chung, Mei 37 5th St.

345678901 Adams, John 54B Hague

456789012 Elam, Mary 123-22 E St.

567890123 Bush, G.W 555 Austin

678901234 Clinton, B. 89 Congress

789012345 Bush, J. 555 Austin

Given:

StudentID Name

123456789 Saenz, Lupe

234567890 Chung, Mei

345678901 Adams, John

456789012 Elam, Mary

567890123 Bush, G.W

678901234 Clinton, B.

789012345 Bush, J.

CIS3465: Professor Kirs Slide Number: 28

RDBMS Basics

Basic RDBMS Operators: JOIN

Links 2 different tables together (Producing a different table)

Each Table MUST have attributes that are common to both tables

123CID

124125126127

CNAABBCCBBDD

CA6

20129

18

CRAABBCC

CAdd123 A234 B345 C

TableA TableB

CIS3465: Professor Kirs Slide Number: 29

RDBMS Basics

Basic RDBMS Operators: JOIN

Consider the relationship between a Physician and his/her Patients

123456789

PhysID

Physician Patient

234567890

345678901

Physname

Smith, Mary

Brown, John

Jones, Jerry

PatID Patname PhysID

987654321 Washington, G. 234567890

876543210

765432109

654321098

543210987

Adams, J.Q.

Jefferson, T.

XXX

YYY

ZZZ

AAA

432109876

321098765

345678901

123456789

123456789

234567890

345678901

123456789

CIS3465: Professor Kirs Slide Number: 30

RDBMS Basics

Basic RDBMS Operators: JOIN

The Product of the tables is:Physician Patient

123456789 Smith, Mary

234567890 Brown, John

345678901 Jones, Jerry

PatnamePatID Phys

Washington, G.987654321 234567890

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

PhysID Physname PhysID Physname PatID Patname Phys

Adams, J.Q.876543210 345678901123456789 Smith, Mary

123456789 Smith, Mary Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

123456789 Smith, Mary Jefferson, T.765432109 123456789

123456789 Smith, Mary XXX654321098 123456789

123456789 Smith, Mary YYY543210987 234567890

123456789 Smith, Mary ZZZ432109876 345678901

=*

234567890

234567890

234567890

234567890

234567890

234567890

Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Brown, John

Brown, John

Brown, John

Brown, John

Brown, John

Brown, John

Washington, G.987654321 234567890

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Adams, J.Q.876543210 345678901

345678901

345678901

345678901

345678901

345678901

345678901

Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Washington, G.987654321 234567890

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Adams, J.Q.876543210 345678901

The product of a 3 (row) x 2 (column) Table and a 6 (row) x 3 (column) Table would be a 18 (row) by 5 (column) Table

CIS3465: Professor Kirs Slide Number: 31

RDBMS Basics

Basic RDBMS Operators: JOIN

Now, entering the command: Select Patname … Where PhysID = Phys would yield:

PhysID Physname PatID Patname Phys

Adams, J.Q.876543210 345678901123456789 Smith, Mary

123456789 Smith, Mary Washington, G.987654321 234567890

123456789 Smith, Mary Jefferson, T.765432109 123456789

123456789 Smith, Mary XXX654321098 123456789

123456789 Smith, Mary YYY543210987 234567890

123456789 Smith, Mary ZZZ432109876 345678901

234567890

234567890

234567890

234567890

234567890

234567890

Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Brown, John

Brown, John

Brown, John

Brown, John

Brown, John

Brown, John

345678901

345678901

345678901

345678901

345678901

345678901

Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

PhysID Physname PatID Patname Phys

Adams, J.Q.876543210 345678901123456789 Smith, Mary

Smith, Mary Washington, G.987654321123456789 234567890

123456789 Smith, Mary Jefferson, T.765432109 123456789

123456789 Smith, Mary XXX654321098 123456789

123456789 Smith, Mary YYY543210987 234567890

123456789 Smith, Mary ZZZ432109876 345678901

234567890

234567890

234567890

234567890

234567890

234567890

Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Brown, John

Brown, John

Brown, John

Brown, John

Brown, John

Brown, John

345678901

345678901

345678901

345678901

345678901

345678901

Washington, G.987654321 234567890

Adams, J.Q.876543210 345678901

Jefferson, T.765432109 123456789

XXX654321098 123456789

YYY543210987 234567890

ZZZ432109876 345678901

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

Jones, Jerry

CIS3465: Professor Kirs Slide Number: 32

RDBMS Basics

Basic RDBMS Operators: JOIN

123CID

124125126127

CNAABBCCBBDD

CA6

20129

18

TableA

CRAABBCC

CAdd123 A234 B345 C

TableB

• Natural Joins (Also called Inner Joins)• ONLY the rows with common values in their

join columns are retained.

TableC

CID CN CA CAdd123 AA 6 123 A

124 BB 20 234 B

125 CC 12 345 C126 BB 9 234 B

123 AA 6 123 A

CIS3465: Professor Kirs Slide Number: 33

RDBMS Basics

How??? A Simple PRODUCT of the Tables is First created

123CID

124125126127

CNAABBCCBBAA

CA6

20129

18

TableA

CRAABBCC

CAdd123 A234 B345 C

TableBCN CR CAddCID CA

123 6 AA 123 AAA124 20 AA 123 ABB125 12 AA 123 ACC126 9 AA 123 ABB127 18 AA 123 AAA123 6 BB 234 BAA124 20 BB 234 BBB125 12 BB 234 BCC126 9 BB 234 BBB127 18 BB 234 BAA123 6 CC 345 CAA124 20 CC 345 CBB125 12 CC 345 CCC126 9 CC 345 CBB127 18 CC 345 CAA

CR AA Joins

CR BB Joins

CR CC Joins

CIS3465: Professor Kirs Slide Number: 34

RDBMS Basics

How??? A Simple PRODUCT of the Tables is First created

CN CR CAddCID CA123 6 AA 123 AAA124 20 AA 123 ABB125 12 AA 123 ACC126 9 AA 123 ABB127 18 AA 123 AAA123 6 BB 234 BAA124 20 BB 234 BBB125 12 BB 234 BCC126 9 BB 234 BBB127 18 BB 234 BAA123 6 CC 345 CAA124 20 CC 345 CBB125 12 CC 345 CCC126 9 CC 345 CBB127 18 CC 345 CAA

A SELECT command where CN = CR is issuedTableC

CID CN CA CAdd123 AA 6 123 A124 BB 20 234 B

125 CC 12 345 C126 BB 9 234 B

CIS3465: Professor Kirs Slide Number: 35

RDBMS Basics

ADDITIONAL RDBMS Operators: UNION

(NOT always available)

May use single single table

E.G., Creating a phone directory from 2 separate tables

There are no common values so a join can not be performed

Separate queries will produce separate tables

Duplicate records NOT displayed

Combines the results of two unrelated queries into a single output

CIS3465: Professor Kirs Slide Number: 36

RDBMS Basics

ADDITIONAL RDBMS Operators: INTERSECT

(NOT always available)

Similar to UNION (Combines the results of two unrelated queries into a single output)

There MUST be matching rows in the tables

E.G., Creating a phone directory of Faculty with offices in BUS who also teach in the Business Bldg. Some Faculty Have offices in BUS Some Faculty Teach in the Business Building Merge those who do both into one listing

CIS3465: Professor Kirs Slide Number: 37

RDBMS Basics

ADDITIONAL RDBMS Operators: DIFFERENCE

(NOT always available)

Yields all rows in one table NOT found in the other

Produces list of all possible pairings

PRODUCT (Seen Earlier)

DIVIDE

CIS3465: Professor Kirs Slide Number: 38

RDBMS Basics

RDBMS Software Classifications Fully Relational

Supports all relational algebra functions Enforces Entity and Referential Integrity Rules

Relationally Complete Supports all relational algebra functions Does NOT support Integrity Rules

Minimally Relational Supports only SELECT, PROJECT, and JOIN

Tabular Supports only SELECT, PROJECT, and JOIN User MUST define all access paths

CIS3465: Professor Kirs Slide Number: 39

RDBMS Basics