Week 10- Abstract Data Types and Encapsulation

Question 1

What are the three types of member clauses in C++ for information hiding?

Answer 1

Private, public, and protected clauses.

Question 1

What are some design issues for ADTs?

Answer 1

Form of the container for the interface.
Parameterization of abstract types.
Access controls provided.
Physical separation of specification from implementation.

Question 1

What is process abstraction?

Answer 1

Process abstraction involves subprograms where nearly all programming languages support it. For example, when sorting a list, the specific sorting algorithm (merge, quick, select, insertion) is abstracted away.

Question 1

What are the advantages of hiding data in ADTs?

Answer 1

Reliability, Reduced Complexity, Reduced Name Conflicts, Modifiability: Client code does not need to change if the implementation changes.

Question 2

What is data abstraction?

Answer 2

Data abstraction involves the use of record types and includes subprograms that manipulate data, with access controls to hide details.

Question 3

How can data members be accessed in ADTs?

Answer 3

Indirectly via getters and setters. Read-only versions can be provided by only offering getters, and setters can enforce constraints.

Question 4

What is a key feature of a stack ADT?

Answer 4

A stack allows access to the data element at the top only, typically using the pop() operation as a getter.

Question 5

How are getter and setter methods used to access data members in C#?

Answer 5

Getter and setter methods in C# are used through properties, allowing implementation of getters and setters without explicit method calls.

Question 5

What is the name of a destructor in C++ and how is it distinguished?

Answer 5

A destructor’s name is the same as the class name, preceded by a tilde (~).

Question 5

What is a property in C#?

Answer 5

A property in C# allows encapsulation of a class’s data members, providing a way to access and modify the private fields using getters and setters without explicit method calls.

Question 6

What is the primary purpose of constructors in C++?

Answer 6

Constructors in C++ are used to initialize the data members; they don’t create objects.

Question 7

How do instances of a class in C++ handle member functions and data members?

Answer 7

All instances of a class share a copy of member functions (methods), but each instance has its own copy of class data members (instance variables).

Question 8

Describe the significance of friend functions or classes in C++.

Answer 8

Friend functions or classes allow access to private members from unrelated units and must be declared inside a class.

Question 9

What is the significance of the StackClass example provided for Java?

Answer 9

It illustrates how to implement a stack using Java, demonstrating constructors, methods for stack operations (push, pop, top, empty), and access control.

Question 10

Provide an example of how to use a property to access an instance variable bar in an instance foo.

Answer 10

a = foo.bar; // getter
foo.bar = 3.5; // setter

Question 11

Differentiate between local variables, instance variables, and class variables in Ruby.

Answer 11

Local variables: regular identifiers.
Instance variables: begin with @.
Class variables: begin with @@.

Question 12

What are the two primary features of Abstract Data Types (ADTs)?

Answer 12

The two primary features of ADTs are the packaging of data with their associated operations and information hiding.

Question 13

What are the three major language features of OOP?

Answer 13

The three major language features of OOP are abstract data types, inheritance, and polymorphism.

Question 14

How does inheritance improve productivity in OOP?

Answer 14

Inheritance allows new classes to be defined in terms of existing ones, enabling reuse of ADTs with minor changes and defining classes in a hierarchy.

Question 15

What are subclasses and how do they relate to parent classes?

Answer 15

Subclasses, or derived classes, inherit from parent (superclass) classes. They can add new variables and methods, modify inherited methods, and hide some inherited entities.

Question 16

What are the two kinds of variables in a class?

Answer 16

The two kinds of variables in a class are class variables (one per class) and instance variables (one per object).

Question 17

What are the two kinds of methods in a class?

Answer 17

The two kinds of methods in a class are class methods (accept messages to the class) and instance methods (accept messages to objects).

Question 18

What is the difference between single inheritance and multiple inheritance?

Answer 18

Single inheritance allows a class to inherit from only one parent class, while multiple inheritance allows a class to inherit from multiple parent classes, which can lead to conflicts and interdependencies among classes.

Question 19

What is dynamic binding and how does it relate to polymorphism?

Answer 19

Dynamic binding allows the method of a subclass to override a parent’s method, with the specific method being called determined at run-time. This supports polymorphism, allowing polymorphic variables to hold objects of a class or any of its descendants.

Question 20

What is an abstract (virtual) method and an abstract class?

Answer 20

An abstract (virtual) method defines a protocol without a definition, and an abstract class includes at least one abstract method and cannot be instantiated.

Question 21

List some of the design issues for OOP languages.

Answer 21

Some design issues for OOP languages include:
Whether everything is an object (exclusivity of objects)
Whether subclasses are subtypes
Whether the language supports single or multiple inheritance
How objects are allocated and deallocated
The use of dynamic and static binding
The presence of nested classes
How objects are initialized

Question 22

What is the advantage of having everything as an object?

Answer 22

The advantage is the elegance, purity, and homogeneity of all data structures.

Question 23

What is the disadvantage of having everything as an object?

Answer 23

The disadvantage is that it can be slow for simple objects.

Question 24

In most object-oriented languages, are subclasses considered subtypes?

Answer 24

Yes, most OO languages consider subclasses as subtypes.

Question 25

What does it mean for a subclass to be a subtype of its parent class?

Answer 25

It means that an “isa” relationship holds between the parent class and subclass, and instances of the subclass must behave similarly to instances of the parent class.

Question 26

What can a subclass do in relation to its parent class?

Answer 26

A subclass can add variables and methods, and override methods in compatible ways.

Question 27

Why are subclasses created according to ontological reasons?

Answer 27

Subclasses are made for ontological reasons, not just for functionality and reuse, to ensure the hierarchy makes sense conceptually (e.g., making an airplane a subclass of a vehicle rather than a bird).

Question 28

What are the advantages of multiple inheritance?

Answer 28

The advantages include convenience (methods and variables from multiple sources) and ontological usefulness (e.g., an aircraft can be both a vehicle and a flying object).

Question 29

What are the disadvantages of multiple inheritance?

Answer 29

The disadvantages include increased complexity in the language and implementation, handling name collisions, and potential inefficiency due to increased cost of dynamic binding.

Question 30

Where can objects be allocated if they are treated as other ADTs?

Answer 30

They can be allocated anywhere, such as the run-time stack or heap.

Question 31

Where are objects allocated if they are heap-dynamic only?

Answer 31

They are allocated on the heap.

Question 32

What are the benefits of allocating objects on the heap?

Answer 32

References can be uniform via pointer/reference variables, simplifying assignment and allowing implicit dereferencing.

Question 33

What issues can arise with stack-dynamic object allocation?

Answer 33

Stack-dynamic allocation can lead to object slicing, where an instance of a subclass may not have enough room allocated when passed to a subroutine expecting an instance of the parent class.

Question 34

Is object deallocation typically explicit or implicit?

Answer 34

Deallocation can be either explicit or implicit, depending on the language.

Question 35

Should all binding of messages to methods be dynamic?

Answer 35

It depends. If none are dynamic, the advantages of dynamic binding are lost. If all are dynamic, it can be inefficient. A design may allow the user to specify.

Question 36

Why might a nested class be useful?

Answer 36

Nested classes are useful to avoid cluttering the object system, prevent name clashes, and keep related classes together, such as defining a Node class within a Tree class.

Question 37

Are objects initialized to values when they are created?

Answer 37

This depends on the language. Initialization can be implicit or explicit.

Question 38

How are parent class members initialized when a subclass object is created?

Answer 38

The initialization process for parent class members when a subclass object is created varies by language and typically involves calling the parent class’s constructor.