Transcript or more

Introduction to Database
Review 2
Crows-Feet Notation for ER Diagrams
 This is an alternative to the diamond representation of
relationships.
 Diamond icons are replaced with lines, simplifying the
ER schema.
Zero or one
One or more
Entity 1
Entity 2
 Intuition



Zero or more Exactly one
(mandatory)
means “Zero”
means “One”
means “or more”
Entity 3
Subclasses And Superclasses
 Grouping of the entities of an entity type into
subgroups (forming an "IS-A" relationship).
 Entities in a superclass are grouped into one or more
subclasses.
 An arbitrary number of levels is permitted in a class
hierarchy.
 An entity in a subclass exists in the superclass also
(recursively).
Subclasses/Superclasses Example
 The Customer entity type has the subclasses
PreferredCustomer and Employee.
 All employees are customers.
Customer
O
Employee
Preferred
Customer
Inheritance Among Classes
 Entities in a subclass inherit the attributes of the
superclass.
 A subclass may have its own attributes (termed the
specific attributes).
 Entities in a subclass may participate in relationships
directly (termed specific relationships), and they may
participate in relationships via their superclass(es).
Inheritance Example
 A PreferredCustomer entity inherits the attributes
Name, Address, CutomerID and Balance from
Customer.
 The PreferredCustomer subclass also has the attribute
DiscountLevel.
Address
Name
Customer
CustomerID
Balance
O
EmployeeID
Discount Level
Employee
Preferred
Customer
6
Schema Design Strategies
 Top-down strategy
 Start out with high-level, abstract concepts and apply step-wise
refinements (e.g., specialization) to add "detail."
 Bottom-up strategy
 Start out with basic concepts and apply refinements (e.g.,
generalization).
 Inside-out strategy
 Start with a few central concepts and successively include
additional concepts.
 Mixed strategy
 Combine the above strategies.
Top-down Strategy
CustomerID
Name
Length
Customer
(0,m)
Checked
Out
(0,n)
RentalPrice
VideoTape
TapeNum
CustomerID
Name
Customer
(0,m)
Checked
Out
(0,n)
Title
Status
(1,1)
VideoTape
Length
Copies
(0,n)
Title
FilmID
RentalPrice
Film
Bottom-up Strategy
 Example: discovering a new generalized entity type and
relating it.
Length
CustomerID
Customer
Name
Name
Employee
(0,m)
(0,m)
Cust
Checked
Out
Emp
Checked
Out
EmployeeID
Length
RentalPrice
(0,n)
VideoTape
(0,n)
Title
Bottom-up Strategy, cont.
 This is converted to:
Length
CustomerID
RentalPrice
Name
Customer
(0,m)
O
Employee
EmployeeID
Preferred
Customer
Checked
Out
(0,n)
VideoTape
Title
Conceptual Database Design Approaches
 Centralized design approach
 Integrate first the requirements for all applications and then
design a single schema.
 Assumes a centralized organization.
 The DBA merges the multiple sets of requirements.
 The DBA designs the schema.
 View integration approach
 Design first a schema for each application in isolation, then
integrate the schemas into a single global schema.
 Each user group can design its own schema.
 The DBA designs the global schema.
Entity Integrity
 Primary Key: A candidate key of a relation is a set of
attributes that satisfy two time independent properties:
 Uniqueness - No two tuples of the relation have the same values
for the set of attributes forming the candidate key.
 Minimality - No attributes can be discarded from the
candidate key without destroying the uniqueness property.
 No component of the Primary Key of a base relation is
allowed to accept nulls.
Foreign key
 A foreign key is an attribute or attribute combination
of one relation R2 whose values are required to match
those of the primary key of relation R1 where R1 and
R2 are not necessarily distinct. Note that a foreign key
and the corresponding primary key should be defined
on the same domain(s).
Employee
Emp#
e1
e2
e3
ename
red
blue
brown
Dept
Worksfordept
d1
Foreign key
d2
Dept
d1
d2
d3
Dname
Pay
Tax
Art
Mapping an EER Schema to Relations
 In a sequence of steps, a set of relations is created.
 Sometimes automated in CASE tools
1.
2.
3.
4.
5.
6.
7.
8.
Regular entity types
Weak entity types
Binary 1:1 relationship types
Binary 1:N relationship types
Binary M:N relationship types
n-ary relationship types
Multi-valued attributes
Superclass/subclass relationship types
1. Entity Type Maps to a Table
 Create a table for each regular entity type.
 One column in table for each simple attribute
 Derived attributes may or may not appear (your choice)
 Table’s primary key is the primary key of the entity type
 Optimization: If there are no attributes other than the
primary key, and if the entity participates totally in a
relationship, then the table can be eliminated.
2. Weak Entity Type Maps to a Table
 Create a table for each weak entity type
 One column for each simple attribute
 Include column(s) for the primary key of each owner entity
type. These columns are foreign keys
 The primary key is the combination of each owner primary
key and the partial key.
3. Mapping 1-1 Relationship Types
 For each 1:1 binary relationship type, extend one of the
tables for a participating entity type.
 Primary key of the other entity type becomes a foreign key in
this table
 It is best to extend a table of an entity type with total
participation
 Add columns for each of the simple attributes of the
relationship type
 Optimization: Perhaps remove the table corresponding
to the other entity type
4. 1-to-Many Relationship Types
 For each regular 1:N binary relationship type, there are
several approaches
 Option 1: Create a separate table for the relationship type
 Three tables result
 Key of relationship table is key of “many” side
 Option 2: If the relationship is total, then extend a table
corresponding to the ‘many’ entity type
 Two tables result (optimization)
 Option 3: If the relationship is not total, extend a table with
nullable attributes (sometimes not allowed for foreign keys)
 Two tables result (optimization)
5. Many-to-Many Relationship Types
 Create a table for each binary M:N relationship type
 The table has columns for
 A column for each primary key attribute in a participating
entity type. These are foreign keys
 A column for each of the simple attribute of the relationship
type
 The primary key of the table is the union of the
primary keys of the participating entity types
6. N-ary Relationship Types
 Create a table for each n-ary (n > 2) relationship type
 Columns in the table are the primary keys of the participating
entity types. (These are foreign keys)
 Also include columns for each simple attribute of the
relationship type
 The primary key of the created table is the union of the
primary keys of the participating entity types
 Optimization: If the relationship type is (1,1) on a side,
it may be possible to remove an entity table, placing its
attributes in the table associated with the relationship
7. Multivalued Attributes
 Create a table for each multivalued attribute
 The table has a column for each simple attribute of the
multivalued attribute
 Add columns for the primary key of the entity or relationship
type to which the attribute belongs. (This is a foreign key)
 The primary key is the combination of all the attributes
 Example:
Director
Film
 Director (FilmID, Name)
FilmID
Outline
 DDL





Creating/altering schema
Data types
Constraints
DataArchitect mapping from a CDM to a PDM
Referential integrity and other assertions
Data Definition in SQL
 Three statements are used to define the schema in SQL.
 CREATE
 DROP
 ALTER
 These statements apply to
 Tables
 Views
 Domains
Create Table
 Specifies a new base table
CREATE TABLE <table name>
[<size>] <column constraint>,
 Columns with




Name
Data type
Column constraints
Default value
 Table constraints
(<column name> <data type>
... <table constraints> );
Referential Integrity
 Referential integrity says “pointed to” information must
exist.
 A foreign key points to data in some relation
Example
 Customer information must exist for a customer to reserve a film
 No CustomerID can be in Reserves and not in Customer
 Can be specified as a column constraint
CREATE TABLE Reserves (...
CustomerID INTEGER
CONSTRAINT
ReservesToCustomerFK REFERENCES Customer(ID),
...)
 Can be specified as a table constraint
CREATE TABLE Reserves (..., CONSTRAINT ReservesToCustomerFK
FOREIGN KEY (CustomerID) REFERENCES Customer(ID)
... )
Referential Integrity Violation Remedies
 Can specify ON UPDATE and ON DELETE options
 Example
CREATE TABLE Reserves (..., CONSTRAINT ResToCusFK
FOREIGN KEY (CustomerID) REFERENCES Customer(ID)
ON DELETE CASCADE ON UPDATE SET NULL
... )
 Options (next slide)
 Note: Child table - has the foreign key, references key in parent
table
 Example: Customer is parent, Reserves is child
Remedy Options
 None
 Update or delete parent value
 No change to matching child value
 Restrict
 Cannot update or delete parent value if one or more matching
values exist in the child table
 No change to matching child value
 Cascade
 Update or delete parent value
 Update or delete matching values in child table
Remedy Options, cont.
 Set null
 Update or delete parent value
 Set matching values in child table to NULL
 Set default
 Update or delete parent value
 Set matching values in child table to default value
Retrieval Queries in SQL: SELECT
 SQL has one basic statement for retrieving information
from a database; the SELECT statement.
 The basic form of the SQL SELECT statement is called
a mapping or a select-from-where block.
SELECT column list
FROM table list
WHERE
condition
Outline - The SELECT statement
 Single table
 Projection
 Selection
 Multiple tables
 Cartesian product and join
 Set operations
 Subqueries
 Optional clauses
 Ordering results
 Computing aggregates on groups
 Additional joins
Modifications
 There are three modification statements
 INSERT
 UPDATE
 DELETE
 For insertions, either values can be specified, or a select
statement provides the values
 Enter a reservation for Eric for the film 332244
INSERT INTO Reserved
VALUES (123456, 332244, CURRENT_DATE)
View
 Views provide a mechanism to create a virtual table
CREATE VIEW name AS query expression
 To create a view we use the command
 Define a view of all customers in Dublin
CREATE VIEW Dublin_Customers AS
SELECT *
FROM Customer
WHERE City = ’Dublin’
View, cont.
 Define a view of all customers holding reservations, and
the films they have reserved
CREATE VIEW Reservations AS
SELECT Name, Title
FROM Customer, Reserved, Film
WHERE Customer.CustomerID = Reserved.CustomerID
AND Reserved.FilmID = Film.FilmID
Transactions
 A transaction can be defined syntactically: each
transaction, irrespective of the language in which it is
written, is enclosed whthin two commands
begin transaction
end transaction
 Within the transaction code, two particular instructions
can appear
commit work
rollback work
Triggers
 The creation of triggers is part of the DDL
 Maintain data integrity
 Associated with a table (view)
 Event-condition-action
 Wait for a table event
before
after
X
insertion
deletion
update
 On event, evaluate condition
 If condition is true, execute action
Potential Applications
 Notification
 an active database may be used to monitor
 Enforce integrity constraints
 Business roles
 Maintenance of derived data
 Maintain the derived attribute whenever individual tuples are
changed
Trigger Gotchas
 Potentially infinite loop
 Trigger A: On insertion into Person, insert into Population
 Trigger B: On insertion into Population, insert into Person
 Mutating tables
 Trigger A: On insertion into Person, insert into Person!
 Disallowed!
 Trigger cannot make changes to table that trigger is defined on
The Object Database
 Object databases integrate database technology with
the object-oriented paradigm
 In object databases, each entity of the real world is
represented by an object. Classical examples of objects
are:
 Electronic components, designed using a Computer Aided
Design (CAD) system;
 Mechanical components, designed using a Computer Aided
Manufacturing (CAM) system;
 Specifications and programs, managed in a Computer Aided
Software Engineering (CASE) environment;
 Multimedia documents, which includes text, images and sound,
managed by multimedia document managers.
Why OODB?
 From programming language point of view:
 permanent storage of objects (languages just support objects in
memory)
 sharing of objects among programs
 fast, expressive queries for accessing data
 version control for evolving classes and multi-person projects
Why OODB?
 From database point of view:
 More expressive data types (traditional DBs provide limited predefined types)
 e.g., a desktop publishing program might model a page as a series of
frames containing text, bitmaps, and charts
 need composite and aggregate data types (e.g., structures and arrays)
 More expressive data relationships
 many-to-one relationship (e.g., many students in one class)
 navigating across relationship links
 More expressive data manipulation
 SQL is relationally complete but not computationally complete i.e., great
for searching for lousy for anything else
– leads to use of conventional programming language plus SQLinterface
– overhead of mapping from SQL to conventional languages
 Better integration with programming languages (esp. OO languages)
 Encapsulation of code with data
40
Two Object-oriented Approaches
 Object-oriented (OODBMS)
 Hellerstein - “to add DBMS capabilities to an O-O language”
 Persistence, object lives beyond program execution
 PJava - persistent Java
 Several commercial products
 Object-relational (ORDBMS)
 Hellerstein - “extends a relational database with O-O features”
 Rich data types
 Inheritance
 Several commercial vendors, SQL3
OODBMS
 Advantages
 Removes impedance mismatch
 Long-lived transactions
 Enriched modeling
 Disadvanatages
 Lack of universal query language
 Lack of agreed upon standard
 Performance depends on class definition