Topics-Ch4Ch5

Download Report

Transcript Topics-Ch4Ch5

Database
Design and
Maintenance
Review
• Proper database design ensures that the data is
represented properly, tables are joined correctly, and
that data can be easily and accurately retrieved.
Relational Database Terms
• Relation = Table
• Tuple = Row
• Attribute = Field
Relational Database
Characteristics
• No two tuples can be exactly the same.
• The order of tuples has no significance.
• Each attribute must describe the relation, and have a unique
name.
• Each attribute can have only one value in a tuple.
• An attribute must have the same set of possible values
(domain) in all tuples.
Building a
Relational Database
•
•
•
•
Designing tables
Creating tables
Joining tables
Designing and creating other
objects
Recognizing Table Types
• Master tables – contain data about people and things.
• Lookup tables – contain data about groups or categories of
information.
• Bridge or Transaction tables – contain data about transactions
and events. Often used to simplify many-to-many joins.
Identifying a Primary Key
• One or more fields that uniquely identify each
record
• Primary key field must not be blank in any record.
Data Dependency
• Functional dependency - when any attribute determines the
value of another attribute.
• Transitive dependency – when a non-key attribute determines
the value of another attribute.
• Partial dependency – when only one field in a multiple-field
primary key determines the value of another attribute.
Well-Structured Relations
What constitutes a well-structured relation? Intuitively, a
well-structured relation contains minimal redundancy and
allows users to insert, modify, and delete rows in a table
without errors or inconsistencies.
EmpID
Name
Dept
Salary
230
Pillsbury
Marketing
58,000
241
Marshall
Finance
68,400
277
Marco
Accounting
66,000
279
Gunston
Marketing
42,400
290
Jaffe
Planning
49,000
EMPLOYEE1 Table
Well-Structured Relations
EMPLOYEE1 is a well-structured relation. Each row of the
table contains data describing one employee, and any
modification of an employee’s data (such as a change in
salary) is confined to one row in the table.
EmpID
Name
Dept
Salary
230
Pillsbury
Marketing
58,000
241
Marshall
Finance
68,400
277
Marco
Accounting
66,000
279
Gunston
Marketing
42,400
290
Jaffe
Planning
49,000
EMPLOYEE1 Table
Well-Structured Relations
In contrast, EMPLOYEE2 is not a well-structured relation. Notice
the redundancy. For example, values for EmpID, Name, Dept, and
Salary appear in two separate rows for employees 241 and 290.
EmpID
Name
Dept
Salary
Course
Date
230
Pillsbury
Marketing
58,000
C++
2/12/06
241
Marshall
Finance
68,400
SPSS
5/30/07
241
Marshall
Finance
68,400
Web Design
11/2/08
277
Marco
Accounting
66,000
C#
12/8/07
279
Gunston
Marketing
42,400
Java
9/10/06
290
Jaffe
Planning
49,000
Tax Acct
4/22/06
290
Jaffe
Planning
49,000
Bus Adm
6/6/08
EMPLOYEE2 Table
Data Anomalies
Redundancies in a table may result in errors or inconsistencies (called
anomalies) when a user attempts to update the data in the table.
There are three types of data anomalies:
• Insertion anomaly – Suppose we need to add a new employee to
EMPLOYEE2. Since the primary key is (EmpID, Course), to insert a row
both EmpID and Course must be supplied. This is an anomaly, because the
user should be able to enter employee data without supplying Course data.
• Deletion anomaly – Suppose that the data for employee 241 are deleted.
This will result in losing information that this employee completed a course
(SPSS) on 5/30/07.
• Modification anomaly – Suppose that employee 290 gets a salary increase.
We must record the increase in each of the rows for that employee;
otherwise the data will be inconsistent.
The problem with relation EMPLOYEE2 is that it contains data about two
entities: EMPLOYEE and COURSE. We will use normalization techniques to split
EMPLOYEE2 into two relations, one for employee data and one for course data.
Normalizing Tables
On the previous four slides we presented an intuitive discussion of wellstructured relations. We need a more formal procedure for designing
them. Normalization is the process of successively reducing
relations with anomalies to produce smaller, well-structured relations.
Some of the goals are:




Minimize data redundancy, thereby avoiding anomalies and
conserving storage space.
Simplify the enforcement of referential integrity constraints.
Make it easier to maintain data (insert, delete, update).
Provide a better design that is an improved representation of the real
world and a stronger basis for future growth.
First Normal Form
• All fields describe the entity represented by the
table.
• All fields contain simplest possible values.
• No multivalued attributes (also called repeating
groups).
1st NORMAL
City
Chicago
State
IL
NOT
Home Town
Chicago, IL
Not 1NF
EmpID
Dept
CourseName
DateCompleted
203
Finance
Tax Accounting
6/22/07
421
Info Systems
Java
Database Mgt
10/7/07
6/4/06
666
Marketing
A multivalued
attribute
Another
multivalued
attribute
1NF - Eliminating multivalued
attributes
EmpID
Dept
CourseName
DateCompleted
203
Finance
Tax Accounting
6/22/07
421
Info Systems
Java
10/7/07
421
Info Systems
Database Mgt
6/4/06
666
Marketing
This new table does have only single-valued attributes
and so satisfies 1NF. However, as we saw, the table still
has some undesirable properties.
Second Normal Form
• Table is in First Normal Form.
• No partial dependencies exist. (No nonkey fields
are determined by only part of a multiple-field
primary key, i.e., nonkeys are identified by the
whole primary key)
2nd NORMAL
Student ID* Course#*
12345
CIS 101
NOT
Grade
B
Student ID* Course#*
12345
CIS 101
Determines
Determines
*Primary key
Name
Higgins
Third Normal Form
• Table is in Second Normal Form.
• No transitive dependencies (no nonkey fields are
determined by other nonkey fields, i.e., nonkeys are
identified by only the primary key).
3rd NORMAL
Course# *
CIS 101
NOT
Textbook
Credits
Intro to CIS 3
Course# *
CIS 101
Textbook
Book Price
Intro to CIS $45.99
Determines
Determines
*Primary key
Fourth and Fifth Normal Form
• Fourth Normal Form – Table is 3NF and has at most
one multivalued dependency. Can produce records
with many blank values.
• Fifth Normal Form – the table cannot be split
into further tables.
Advanced Field Properties
• Lookup fields
• Multiple-field primary
keys
• Indexes
Lookup Field
• Looks up a value in a joined table.
• Specify “Lookup wizard” in Data Type list.
• Creates an editable query.
Using a Lookup Field
Looks up data values from another table
(or you can create your own list).
Order Customer ID
1008 Smith,Ben
S349
S349
Orders
Customer ID
Name
S349
Smith,Ben
Looks up Name
in Customer table
Customers
Multiple-field Primary Keys
• Also called compound keys or composite keys
• A value in one field in the key can be repeated in
multiple records, but not in all fields of the primary
key.
Student ID* Course# *
12345
CIS 101
Grade
A
12345
B
CIS 200
*Primary key
Index
• Field property that increases
search speed.
• Speeds up sorting and searching
in Datasheet view and all database
objects.
Joining Tables
• One-to-Many join is the most common.
• Other join types:
– One-to-One
– Many-to-Many
One to Many
Table 1
Table 2
Rec. 1
Rec. 1
Rec. 2
One to One
Many to Many
Rec. 1
Rec. 1
Rec. 2
Rec. 1
Rec. 1
Rec. 2
Join Types
• Inner Join
• Left Outer Join
• Right Outer Join
Join types are discussed in the slides to follow,
but also see the discussion at Join-queries.htm.
Inner Join
The default type - includes records with
corresponding values in both tables.

LASTNAME
Gray
Gray
Cedarman
Malderer
Nale
FIRSTNAME
Eric
Nadine
YvonneHRS
Kevin
Rusty
DEPARTMENT
CIS
CIS
HRS
HRS
HRS
HRS
HRS
Only red
records are
included in join.
No department for Nale
DEPARTMENT CODE
1 CIS
HRS
WHS
NAME
Computer Information Systems
Human Resources
Warehouse
No Warehouse employees
Left Outer Join
Includes all records from One table and
corresponding records from Many table

LASTNAME
Gray
Gray
Cedarman
Malderer
Nale
FIRSTNAME
Eric
Nadine
YvonneHRS
Kevin
Rusty
DEPARTMENT CODE
1 CIS
HRS
WHS
DEPARTMENT
CIS
HRS
HRS
Rusty Nale not
included – no
department
assigned.
NAME
Computer Information Systems
Human Resources
Warehouse
How to Change the Join Properties in Query Design View
Double-click on the join line that connects the tables.
Click the bullet next to the join property desired.
For example, in the 240students.mdb database, suppose we
want all students and the pets they own. All students should be
listed regardless of whether they own a pet or not.
This is perfect for a Left Outer Join, because it will select a
student from tblStudents (the “One” table) even if there isn’t a
match (between ID and OwnerID) in tblPets (the “Many” table).
Here’s the SQL and the result of running the query
SELECT ID, FirstName, Name, Breed
FROM tblStudents LEFT JOIN tblPets
ON tblStudents.ID = tblPets.OwnerID
• The result of running this query (in the
240students.mdb database) Left-Join.htm.
Right Outer Join
Includes all records from Many table and
corresponding records from One table.

LASTNAME
Gray
Gray
Cedarman
Malderer
Nale
FIRSTNAME
Eric
Nadine
YvonneHRS
Kevin
Rusty
DEPARTMENT CODE
1 CIS
HRS
WHS
DEPARTMENT
CIS
HRS
HRS
Warehouse
dept. not
included – no
employees
assigned.
NAME
Computer Information Systems
Human Resources
Warehouse
Join Types
Joins displayed as Venn Diagrams
A
A
B
Inner Join
A B
A
B
Left Outer Join
A is “one” table
B
Right Outer Join
B is “many” table
U
A and B are tables
Green striped area is join dynaset.
Join Types
Selected by double clicking on join line between two
tables in Relationship window, and clicking Join Type
button.
Referential Integrity
• Referential integrity keeps the relationships between
tables valid.
• All foreign keys have values that correspond to
records in the referenced table
• Maintain referential integrity by:
– Updating and deleting records when matching
records in a joined table are updated and deleted.
– Eliminating unmatched and duplicated records in
joined tables.
Enforcing Referential
Integrity
•
•
•
•
•
Normalize tables
Set field properties
Use lookup fields
Select specific join type settings
Create and run Find Duplicate and Find Unmatched
queries
Find Duplicate Records Query
• Locates records with duplicate values in Many
table.
Duplicates
One table
Many table
Find Unmatched
Records Query
Locates records in Many table not associated with
record in One table
No match
One table
Many table
Maintaining Databases
•
•
•
•
•
Older versions of Access can be converted to newer versions, and
vice versa.
Databases can be compacted and repaired using the Tools,
Database Utilities command.
Databases can be split into two databases: data and objects
(back and front end).
Databases can be documented using the Tools, Analyze,
Documenter command.
Database performance can be analyzed using the Tools, Analyze,
Performance command.
Object Groups
• Named set of shortcuts that point to database
objects
• Grouped objects are listed together in a single
window.
Modifying Access
Environment
• Tools, Options command allows
changes to the behavior of Access.
• Access standard toolbars can be
modified.