Learn C# Marathi Transliteration

सी शाप शकुया अगद सोपं भाषेत

Get the gist of the concept Write the program See the output for yourself

कौसर चुनावाला

1|Page

2008

Learn C# - do it yourself way...

संगणकशााया व ा!या"साठ$

These notes were compiled by me, while serving NIIT Training Institute, Mumbai – India as a faculty member. This text would facilitate the learning process of the students when they learn an Object-Oriented Programming language like C#. It starts from scratch and takes you upto the depth required at an undergraduate school. I appreciate the contribution made by my students – Farzan, Mario, Swapnil, Geetashree, Swati, Pooja, Shraddha, Chaitali, Snehal, Sushil, Anand, Sadiq, Rama, Tausif, Ameya and all other students and colleagues. Program Listing Chapter 2|Page

Name

Pages

Learn C# - do it yourself way...

1 2 3 4 5 6 7 8 9 10 11 12

3|Page

Introduction to C# C# Fundamentals Decision Making and Looping Constructs Classes, Objects and Methods Collections in C# Polymorphism, Constructors and Destructors Inheritance – Our objects’ family tree Advanced C# Language Concepts File Input and Output Exception Handling Threads Attributes and Reflections

2008

1 Introduction to C# 1.1 Hello World C# Style 1. Open the Notepad and type the following program.

2. Save the file as Hello.cs

3. Go to Visual Studio Command Prompt from Start-> All Programs - >Visual Studio 2008 -> Visual Studio Tools

4. On the command prompt, go to the directory, where you saved the file.

5. Type cs Hello.cs to compile the program.

6. Type Hello to run the program.

1|Page

using System; class Greeting { public static void Main(string[] args) { Console.WriteLine("Hello World"); } }

Output :

Learn C# - do it yourself way...

2008

Dissecting the Hello World Program using System; class Greeting { public static void Main(string[] args) { Console.WriteLine("Hello World"); } }

The program starts with the using statement. Wait for a while, till I tell you what it means. Then, we declare a class called Greeting. All code in C# must be written inside a class. Next, we have written Main(). The word Main() indicates this is the start of the program. This is where the C# system start the executing(running) the program. Thus, Main() is the entry point(starting point) of any program. All statements written inside Main() are executed line by line. The Main() block is started using a starting curly brace { and terminated using an ending curly brace }. Whatever you want to write inside Main() must be enclosed within this pair of curly braces. On the same lines, whatever you write inside the class, must be enclosed within { ... }. In the above program, we have written just one command/statement – Console.WriteLine().This command is used display text on the VDU screen. Whatever you put inside Console.WriteLine() gets printed on the screen, when you run the program. In this case, it is Hello World.

2|Page

Learn C# - do it yourself way...

2008

What are Classes and Objects? C# is an object-oriented programming language. All the C# code you write will appear in a class. In early days of programming, programs were designed using flowcharts and a sort of top-down design. With this type of design, a large problem was solved step-by-step. Thus, the solution to a problem was visualised as a series of steps or operations. To arrive at the solution, you follow the sequence of steps or procedure. This form of programming is called Procedural Programming. The new paradigm of programming is Object-Oriented Programming. In Object-Oriented Programming, to find the solution, we decompose a large problem in terms of objects. So, what really are objects. Any real world entity or thing, say for example a car, a person, a flight, a ticket-booking, an employee, a college can be an object in C#. To model real world things, in C# we create objects. Just like in the real – world, each object has some features or characteristics that describe the object, our C# objects also have features that describe the object. For example, every car has some name, brand, year of manufacture, price, color which describe it. In C#, to model this concept, C# car object will also have attributes name, brand, yearOfManufacture, price, color etc. Every car object will have its own values of these attributes. Consider the following two car objects –

3|Page

2008

Learn C# - do it yourself way...

6 Polymorphism, Constructors and Destructors 6.1 Function Overloading There are 6 different versions of the Multiply() method in the Test class. st The 1 version multiplies two ints and returns an int as the answer. nd The 2 version accepts 2 doubles as input parameters and returns a rd double result as the answer. The 3 version accepts an integer and a double number as an argument, and returns their product as the answer. Every time, we are assigning a new extra meaning to the Multiply() function. Thus, we have overloaded the Multiply() method.

class Test { public int Multiply(int a, int b) { return a * b; } public double Multiply(double a, double b) { return a * b; } public double Multiply(int a, double b) { return a * b; } public double Multiply(double a, int b) { return a * b; } public int Multiply(int a, int b, int c) { return a * b * c; } } class OverloadDemo { public static void Main() { Test t = new Test(); double res1 = t.Multiply(4.2, 5.5); int res2 = t.Multiply(3, 5); double res3 = t.Multiply(4.2, 5);

4|Page

Learn C# - do it yourself way...

2008

double res4 = t.Multiply(5, 3.8); int res5 = t.Multiply(3, 4, 5); }

}

In method overloading, we assign a new extra meaning to the method. For example suppose, we have written the Multiply() method as follows : public int Multiply(int x, int y) { int result = x * y; return result; } The Multiply() method is designed to accept 2 integers as input parameters. It multiplies the two integers x and y, and returns an integer result as the answer. Suppose, we now define one more Multiply() method as follows public double Multiple(double x, double y) { double result = x * y; return result; } This Multiply() method multiplies two double numbers and returns a double result as the answer. Thus, the Multiply() method now has two different versions. The first version can multiply two integers, whereas the second version can multiply 2 doubles. Thus, we have assigned a new extra meaning to the Multiply() method, so that it can now no longer multiply just two ints, but it can as well multiply two doubles.Hence, we have overloaded the Multiply() method. Thus, this is method overloading. Let us define one more Multiply() method as follows 5|Page

Learn C# - do it yourself way...

2008

public double Multiply(int x, double y){ double result; result = x * y; return result; } This time round, the Multiply() method can multiply one int with one double value, and returns a double result as the answer. Thus, we are again assigning a new extra meaning to the Multiply() method. So, once again we have overloaded the Multiply() function. If we define one more Multiply() method as : public double Multiply(double x, int y) { double result; result = x * y; return result; } This new version of the Multiply() method can now multiply a double with an integer. Mutiplying a double with an int, is different from multiplying an int with a double. Thus, we are again assigining a new meaning to the Multiply() method. Hence, we have again overloaded the Multiply() method. Consider another version of the Multiply method public int Multiply(int x, int y, int z){ return x * y * z; } This version of the Multiply() method computes the product of 3 ints instead of 2. Once again we have assigned extra meaning to Multiply(). So, it is an overloaded function. A function is said to be overloaded if 6|Page

2008

Learn C# - do it yourself way...

1. When two or more methods in the same class have the same name 2. They have different Parameter Lists Parameter list is different if - Type of parameters differs. - No of parameters is different. - Order of parameters is different.

6.2 Operator Overloading A Date class is used to represent the date in the form of days, months and years.

class Date { public int day; public int month; public int year; public static Date operator +(Date d1, Date d2) { Date result = new Date();

The + operator is overloaded, so that it can add two Date class objects. The two Date objects are passed as arguments into d1 and d2. To store the sum of Date objects, we take a result Date object. We set the day, month and year of the result Date object as the sum of the corresponding day, month and year of d1 and d2.

result.day = d1.day + d2.day; result.month = d1.month + d2.month; result.year = d1.year + d2.year; if(result.day>30){ result.day -= 30; result.month++; } if (result.month > 12) { result.month -= 12; result.year++; }

We also need to check for overflow. Overflow happens when the no of resulting days exceeds 30. For example, we treat 35 days as 1 month and 5 days. So, when an overflow occurs we add 1 to the months, and the balance days are 35 – 30 = 5.

The < operator is overloaded, so that it can compare 2 Date objects and find out which is smaller. Comparing 2 Date objects for the no. Of days, months and years can be tricky, so a simple way would be to find the total no of days of each Date object and then just compare them. If days1 turns out to be less than days2 as assumed, return true else return false. We also overload > operator, since they must be overloaded in pairs.

7|Page

return result; } public static bool operator <(Date d1, Date d2) { bool ans; int days1 = d1.day + d1.month * 30 + d1.year * 365; int days2 = d2.day + d2.month * 30 + d2.year * 365; if (days1 < days2) ans = true; else ans = false; return ans; } public static bool operator >(Date d1, Date d2)

Learn C# - do it yourself way... { bool ans; int days1 = d1.day + d1.month * 30 + d1.year * 365; int days2 = d2.day + d2.month * 30 + d2.year * 365; if (days1 > days2) ans = true; else ans = false; ; return ans; } public void GetDate() { Console.WriteLine( .WriteLine("Day : " + day); Console.WriteLine( .WriteLine("Month : " + month); Console.WriteLine( .WriteLine("Year : " + year); Console.WriteLine(); .WriteLine(); } } class DateDemo { public static void Main() { Date d1 = new Date(); Date Date d2 = new Date(); Date Date result; bool ans; d1.day = 2; d1.month = 7; d1.year = 5; d2.day = 29; d2.month = 6; d2.year = 4; result = d1 + d2; d1.GetDate(); d2.GetDate(); result.GetDate(); ans s = (d1 > d2); Console.WriteLine( .WriteLine("d1 > d2 : " + ans); } }

Suppose we write the instruction 2 + 3.

8|Page

2008

Learn C# - do it yourself way...

2008

When the C# system encounters this line, it does not directly add the 2 + 3. Instead, the C# system calls a function with the name + and passes 2 and 3 as the arguments. +(2,3) When this function is called, the control jumps to the function definition. The + function looks like this public int operator +(int a, int b) { ___________________; ___________________; ___________________; } Since, the computer gives us the answer of 2 + 3 = 5, this function + is already defined in C#. Now, suppose we would like to assign extra meaning to the + operator. Let's say we have the following objects MyClass obj1 = new MyClass(); MyClass obj2 = new MyClass(); Suppose, we now write obj1 + obj2 Thus, the following function is called +(obj1,obj2) But, since these our own objects, we have created them, we have made their class design, the + operator does not know how to add these custom made objects. To be able to add them, we need to define the following function -

9|Page

Learn C# - do it yourself way...

2008

public MyClass operator + (MyClass obj1, MyClass obj2){ MyClass result; //Code for adding the 2 objects.. _______________________________; _______________________________; return result; } By defining this function, the + operator will also be able to add objects. Hence, we are assigning extra meaning to the + operator, so that it is able to add our own objects. Hence, it is called Operator Overloading. Syntax for Overloading Binary Arithmetic Operator public static MyClass operator + (MyClass obj1, MyClass obj2) { MyClass result; ______________ ______________ return result; } Syntax for Overloading Unary Arithmetic Operator public static MyClass operator -(MyClass obj) { MyClass result; ______________ ______________ return result; }

10 | P a g e

Learn C# - do it yourself way...

2008

Syntax for Overloading Relational Operators public static bool operator <(MyClass obj1, MyClass obj2){ bool ans; //Compare the two objects ______________ ______________ return ans; } Note : Relational operators are always overloaded in pairs. If we overload < operator, then we must also overload > operator. Similarly, if we overload <= or == operator, we must also overload their corresponding complementary operators >= and !=.

6.3 Constructors class Box { double width; double height; double depth; public Box() { width = height = depth = 1; } public Box(int l) { width = height = depth = l; } public Box(int w, int h, int d) { width = w; height = h; depth = d; } public void GetBox() { Console.WriteLine("w : " + width + " h : " + height + " d : "+depth); }

11 | P a g e

Learn C# - do it yourself way...

2008

public void Volume() { Console.WriteLine( .WriteLine("Volume : " + width * height * depth); } } class BoxDemo { public static void Main() { Box myBox1 = new Box(); Box myBox2 = new Box(5); Box myBox3 = new Box(10, 20, 30); myBox1.GetBox(); myBox2.GetBox(); myBox3.GetBox(); myBox1.Volume(); myBox2.Volume(); myBox3.Volume(); } }

A constructor is a special method that is called automatically, when an object is created.

Why Constructors? When we first create an object, all its fields are initialised to 0. In other words, the memory is zeroed out.

Box myBox; myBox = new Box();

To be able to use this Box object, we must put in meaningful values in the width, height and depth of myBox. Thus, we must assign values to myBox’s instance variables. To do so, we must manually go in and initialise all the fields of the myBox object. 12 | P a g e

Learn C# - do it yourself way...

2008

myBox.width = 10; myBox.height = 20; myBox.depth = 30; rosy till we have to work with just 1 Box All of this seems rosy--rosy object. But, consider a real-time real time shipping application, where we to keep track of thousands of orders and consignments. Each consignment ignment is shipped in Box object. So, if we typically have 1000 Box objects; myBox1, myBox2, myBox3,... this process of manually initialising all the fields becomes very tedious and cumbersome.

Well, you can use convenience methods like SetBox(). But, Bu when you have a thousand objects, you have explicitly call/invoke SetBox() method on each Box object. Once again, it calls for a lot of work.

How can Constructors help me?

13 | P a g e

Learn C# - do it yourself way...

2008

Constructor is a special method that is invoked automatically, when an object is created. So, you don’t have to call a constructor method explicitly. Thus, constructor can be used to automatically initialise an object, right at the time when the object is born/created. 1. A constructor has the same name as the class. 2. Like methods, constructors can also accept input parameters. 3. Unlike methods, constructors do not have a return type, not even void. 4. Like methods, constructors can also be overloaded. Depending upon the call, different versions of the constructor will be invoked. Note – Adding constructors to a class is compulsory. Q. But, the classes that we have written till now seemed to work fine without a constructor. A. If you do not write your own constructor, the C# Compiler adds a constructor method for you for free. Such a constructor is called the default constructor. Q. But how does the default constructor know, what I want it to do? A. It does not! That’s why, it does not accept any parameters and has empty body. It looks like this – public Box() { } Thus, the free constructor that C# compiler adds for you, is a donothing constructor.

14 | P a g e

Learn C# - do it yourself way...

2008

6.4 Destructors, Finalize() and Dispose() We know that when, objects are created, constructors are called, but when objects are destroyed what happens? Unlike C C++, in C# you don’t have to manually destroy objects. Objects which are no longer needed are automatically destroyed by the Garbage Collector(GC). In C#, you can have two kinds of resources in the programs you write. The Objects of a class, primitive data-types etc. are supported by C#, and the space occupied by them is automatically release by the GC, when it finds they are no longer in use. Such resources are called Managed Resources. But, there are other resources, like a Database Connection, a File Stream, a Network Stream, a pipe etc. which are not supported by C#. This means, when they are no longer in use, the responsibility of their clean-up does not lie with the GC. You have to make sure, you clean up these resources. Such resources are called unmanaged resources. If your program/class uses unmanaged resources, you can clean up these resources in two ways – 1)Write a destructor or finalizer method. 2)Override the Dispose() method. If you write a destructor in the class, it is called just before the objects of this class are destroyed by GC. There are two ways to write a destructor – class MyClass { ~MyClass() { Console.WriteLine("You can free up an unmanaged resources here."); } }

15 | P a g e

Learn C# - do it yourself way...

2008

class DestroyDemo { public static void Main() { MyClass obj = new MyClass(); obj = null; } }

A destructor has the same name as the class, but prefixed with a ~ tilde sign. Internally, this Destructor is translated to a call to the Finalize() method – class MyClass { protected override void Finalize() { try { Console.WriteLine("You can free up any unmanaged resources here"); } finally { base.Finalize(); } } }

Now, when you have a number of objects, the order in which these objects will be destroyed/finalized is not fixed or known. Say, if we have five objects obj1, obj2, obj3, obj4, obj5 of a class MyClass. The GC may decide to destroy the objects in any random order say, for example obj3, obj5,obj1, obj4, obj2. The GC maintains a queue of all the objects that are to be finalized. This is called Finalization Queue. In other words, finalization is non-deterministic(random). When we want to exercise control over the finalization process, we can override the Dispose() method (whose interface is IDisposable). The Dispose() method differs from the Finalize() method, in that, we need to explicitly call the Dispose() method. A finalizer is implicitly called, and cannot be explicitly called, even if you wanted to. Implicit Resource Cleanup Explicit Resource Cleanup

16 | P a g e

Destructor or Finalize() method Dispose()

Learn C# - do it yourself way...

2008

7 Inheritance – Our objects’ family tree 7.1 Overview of Inheritance Kathy is assigned the task of building an Employee Management System. She starts by thinking, “I am gonna take Employee objects for every Employee” in the organisation. So, she begins by writing an Employee class as follows – class Employee { public string name; public string address; public int SSN; public int number; public float salary; public float computePay() { return salary / 12.0F; } }

The design of an Employee class seems fine to her initially. An employee has a name, address, and number and so on... We want the compute the pay of different Employee objects. But, Kathy questions herself, does every Employee have a salary? Is it true that every Employee object has a salary. By studying the problem 17 | P a g e

Learn C# - do it yourself way...

2008

domain, she finds out that there are Employees which are paid on an hourly basis, or employees which are on contract basis. The first mistake Kathy made was to add a field of type salary to Employee. She discovers although employees are the objects in our problem domain, there are actually two different types of Employee objects : Salaried Employees and Hourly Employees. Therefore, we should write two classes : Salary and Hourly. The Salary class should have a field to represent the employee’s annual salary because a Salaried Employee has a salary. The Hourly class should have fields to represent the employee’s hourly pay rate and the no of hours for which he worked. The Salary and Hourly classes look like this – class Salary { public string name; public string address; public int SSN; public int number; public int salary; public float computePay() { return salary / 12.0F; } } class Hourly { public string name; public string address; public int SSN; public int number; public int hoursWorked; public int hourlyRate; public int computePay() { return hoursWorked*hourlyRate; } }

Although Salary and Hourly classes are different types, they are not entirely different. In fact, the two types of employees have a lot in common, as seen by the repetition of fields and methods in these two 18 | P a g e

Learn C# - do it yourself way...

2008

classes. So, we can take the common elements from both the classes, and put them in a parent class leaving the unique elements behind in the child class. We can simply make the Salary and Hourly classes inherit the elements of the Employee class. Employee class is called Parent class/Base class/Super class. Salary and Hourly classes are called child class/Derived class/subclass. If you want to make Salary and Hourly the child classes of Employee, we write them as follows : class Employee { public string name; public string address; public int SSN; public int number; } class Salary : Employee { public int salary; public float computePay() { return salary / 12.0F; } } class Hourly : Employee { public int hoursWorked; public int hourlyRate; public int computePay() { return hoursWorked*hourlyRate; } }

When your classes use inheritance, you only need to write your code once.

In the above scenario, Salary class and the Hourly class have a lot of same code.

19 | P a g e

2008

Learn C# - do it yourself way...

Salary Employee and Hourly Employee are both employees. When you have two classes which are more specific cases of something more general, you can set them up to inherit from the same base class. Employee name address SSN number GetEmployee()

Salary salary

Hourly hourlyRate hoursWorked

computePay() computePay()

Build up your class model, by starting General and getting more Specific Specif

7.2 Method Overriding class Employee { public string name; public string address; public int SSN; public int number;

20 | P a g e

Learn C# - do it yourself way... public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine(); } } class Salary : Employee { public int salary; public float computePay() { return salary / 12.0F; } public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine("Salary : " + salary); Console.WriteLine(); } } class Hourly : Employee { public int hoursWorked; public int hourlyRate; public int computePay() { return hoursWorked * hourlyRate; } public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine("Hours Worked : " + hoursWorked); Console.WriteLine("Hourly Rate : " + hourlyRate); Console.WriteLine(); } } class EmployeeDemo { public static void Main() { Employee e = new Employee(); e.name = "Robert Smith"; e.address = "111 Palm street"; e.SSN = 999901111; e.number = 1;

21 | P a g e

2008

Learn C# - do it yourself way...

Salary s = new Salary(); s.name = "Jane Smith"; s.address = "Oak Drive"; s.SSN = 111009999; s.number = 2; s.salary = 10000; Hourly h = new Hourly(); h.name = "George Washington"; h.address = "333 Espresso Lane"; h.SSN = 111990000; h.number = 3; h.hoursWorked = 200; h.hourlyRate = 30; e.GetEmployee(); s.GetEmployee(); h.GetEmployee(); } }

7.3 Use of base keyword using System; class Employee { public string name; public string address; public int SSN; public int number; public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine(); } } class Salary : Employee { public int salary; public float computePay() { return salary / 12.0F; } public void GetEmployee() { base.GetEmployee(); Console.WriteLine("Salary : " + salary); Console.WriteLine(); } }

22 | P a g e

2008

Learn C# - do it yourself way... class Hourly : Employee { public int hoursWorked; public int hourlyRate; public int computePay() { return hoursWorked * hourlyRate; } public void GetEmployee() { base.GetEmployee(); Console.WriteLine("Hours Worked : " + hoursWorked); Console.WriteLine("Hourly Rate : " + hourlyRate); Console.WriteLine(); } } class EmployeeDemo { public static void Main() { Employee e = new Employee(); e.name = "Robert Smith"; e.address = "111 Palm street"; e.SSN = 999901111; e.number = 1; Salary s = new Salary(); s.name = "Jane Smith"; s.address = "Oak Drive"; s.SSN = 111009999; s.number = 2; s.salary = 10000; Hourly h = new Hourly(); h.name = "George Washington"; h.address = "333 Espresso Lane"; h.SSN = 111990000; h.number = 3; h.hoursWorked = 200; h.hourlyRate = 30; e.GetEmployee(); s.GetEmployee(); h.GetEmployee(); } }

7.4 Adding Constructors using System; public class Employee { public string name; public string address; public int SSN;

23 | P a g e

2008

Learn C# - do it yourself way...

2008

public int number; public Employee() { Console.WriteLine("Inside Grandparent"); } public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine(); } } public class Salary : Employee { public int salary; public Salary() { Console.WriteLine("Inside Parent"); } public float computePay() { return salary / 12.0F; } public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine("Salary : " + salary); Console.WriteLine(); } } public class PartTimeSalary : Salary { public PartTimeSalary() { Console.WriteLine("Inside PartTimeSalary"); } } class EmployeeDemo { public static void Main() { PartTimeSalary pts = new PartTimeSalary(); } }

7.5 Calling Parent class Constructors 24 | P a g e

Learn C# - do it yourself way...

2008

using System; public class Employee { public string name; public string address; public int SSN; public int number; public Employee(string n, string a, int S, int num) { name = n; address = a; SSN = S; number = num; } public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine(); } } public class Salary : Employee { public int salary; public Salary(string n,string a, int S, int num, int sal):base(n,a,S,num) { salary = sal; } public float computePay() { return salary / 12.0F; } public void GetEmployee() { Console.WriteLine("Name : " + name); Console.WriteLine("Address : " + address); Console.WriteLine("SSN : " + SSN); Console.WriteLine("Number : " + number); Console.WriteLine("Salary : " + salary); Console.WriteLine(); } } class EmployeeDemo { public static void Main() { Salary s = new Salary("Jane Smith", "222 Oak Drive", 111009999, 3, 10000); s.GetEmployee(); } }

25 | P a g e

Learn C# - do it yourself way...

2008

8 Advanced C# Language Concepts 8.1 Using Parent class references to Child Objects An object reference can refer to an object of its class, or to an object of any class derived from it by Inheritance. Suppose, we have the following class hierarchy.

For example, if Animal class is the parent, and Bird is its child, SongBird is the child of Bird and so on,... Animal reference can refer to a Bird object or a SongBird object. Animal a; 26 | P a g e

Learn C# - do it yourself way...

2008

a = new Animal(); a = new Bird(); a = new SongBird(); Assigning an object to an ancestor reference is considered to be a widening conversion, and can be performed using simple assignment. Animal a = new MockingBird(); Assigning an ancestor object to a child reference can also be done, but it is considered to be a narrowing conversion and must be done with a cast. MockingBird m = (MockingBird)new Bird(); The widening conversion is the most useful for implementing polymorphism.

8.2 Polymorphism via Inheritance A polymorphic reference is the one which can refer to different types of objects at different times. Such an object reference can refer to one object at one time, and can refer to another object(related by inheritance) at another time. It is the type of object being reference, not the reference type, which determines which method is invoked. Polymorphic references are therefore resolved at run-time, not during compilation; hence it is called dynamic binding. class Shape { public virtual void Area() { } }

27 | P a g e

Learn C# - do it yourself way... class TwoDimShape : Shape { public int a; public int b; } class ThreeDimShape : Shape { public int a; public int b; public int c; } class Rectangle : TwoDimShape { public override void Area() { int area = a * b; Console.WriteLine("Area : " + area); } public void SetRectangle(int x, int y) { a = x; b = y; } } class Triangle : TwoDimShape { public override void Area() { double area = 0.5 * a * b; Console.WriteLine("Area : " + area); } public void SetTriangle(int bas, int height){ a = bas; b = height; } } class Cube : ThreeDimShape { public override void Area() { int area = 6 * a * a; Console.WriteLine("Area : " + area); } public void SetCube(int s) { a = b = c = s; } } class Box : ThreeDimShape { public override void Area() { int area = 2 * a * b + 2 * b * c + 2 * a * c; Console.WriteLine("Area : " + area);

28 | P a g e

2008

Learn C# - do it yourself way... } public void SetBox(int x, int y, int z) { a = x; b = y; c = z; } } class ShapeDemo { public static void Main() { int choice; Shape s = new Shape(); Triangle t = new Triangle(); t.SetTriangle(5,10); Rectangle r = new Rectangle(); r.SetRectangle(10,10); Cube c = new Cube(); c.SetCube(5); Box b = new Box(); b.SetBox(10,20,30); while (true) { Console.WriteLine("Enter a choice : "); Console.WriteLine("1 - Rectangle"); Console.WriteLine("2 - Triangle"); Console.WriteLine("3 - Cube"); Console.WriteLine("4 - Box"); Console.WriteLine("5 - Exit"); choice = Convert.ToInt32(Console.ReadLine()); if (choice == 1) s = r; if (choice == 2) s = t; if (choice == 3) s = c; if (choice == 4) s = b; if (choice == 5) break; s.Area(); //Call is resolved at run-time } } }

29 | P a g e

2008

Learn C# - do it yourself way...

2008

8.3 Abstract Classes and Abstract Methods Abstraction is the ability to make a class/method/property abstract in C#. An abstract class is the one which cannot be instantiated. All other functionalities of the class still exist, its fields, its methods and constructors are all accessed in the same way. You just cannot create objects of an abstract class. An abstract class might initially seem like an odd design. Why write a class and not allow anyone to create objects of it? The Employee class that we have written in the previous programs is an example where we do not want to create objects of Employee. Notice, that the Employee class does not have any information about salary and how much his pay comes to. It is safe to say, that no employee of our organisation would like to be strictly an object of Employee. This means, although we need the Employee class as a parent of Salary and Hourly to support inheritance, we do not want to create any objects of this class. We can make it, so that no one can create objects of Employee class, by declaring the Employee class abstract. The only result of making the Employee class abstract is that we can no longer create objects of Employee. If the Employee class is made abstract, the following statement will not compile – Employee e; e = new Employee(“George W.”,”Houston”,43); //Error Just like you can make a class abstract, you can also make a method abstract. For example, we could have a computePay() method in the Employee class as abstract.

30 | P a g e

Learn C# - do it yourself way...

2008

An abstract method is never called. Think about it. If computePay() is abstract in the Employee() class, we really don’t care what computePay() does in the Employee class, because it will never be invoked/called. So, the abstract computePay() method in the Employee class will be empty. Our assumption is that the child classes of Employee will override the computePay(). This is where abstract methods are useful. If you want a class to contain a particular method, but you want the actual implementation to be determined by the child classes, you must declare the method in the parent class as abstract. Abstract methods consist of a method signature, but no method body. public abstract class Employee { public string name; public string address; public int SSN; public int number; public abstract void computePay(); }

8.4 Sealed Keyword Sealed keyword works exactly opposite to abstract. - A Sealed class cannot be sub-classed. - A Sealed method cannot be overridden.

8.5 Namespaces Every class belongs to a namespace. Namespaces basically have two purposes – 1. Namespace provides a mechanism for organising classes. 2. Namespace does compartmentalization. When developing a C# program, you put classes that go together in the same namespace. For example, in C# the classes that are used to 31 | P a g e

Learn C# - do it yourself way...

2008

perform basic Input and Output are in System namespace. The classes used for GUI applications are in System.Windows.Forms namespace. The classes used for creating threads are in System.Threading namespace. Q. Why are namespaces necessary? What if I have a small program that is only a dozen classes? Namespace is more than just a mechanism for organising classes. A more important aspect of namespaces is that they create compartments. For example, suppose that your program contains a class named Vehicle. What if I wrote a class Vehicle as well? How would you be able to tell them apart. What if someone wanted to use my Vehicle class and your Vehicle class in their program? Q. I have seen this problem before. Why don’t you change the names of the classes, such as Vehicle1 and Vehicle2? No thanks. I would have to re-write and re-compile a bunch of code. With namespaces, I don’t have to worry about it. If the two Vehicle classes are in different namespace/compartments, my C# program can distinguish between the two Vehicle classes.

8.6 Adding a class to a Namespace To add a class to a namespace/compartment, we use the namespace keyword. namespace A{ ... } For example, if you wanted to put Employee, Salary and Hourly classes in the payroll namespace/compartment we can do it as follows –

32 | P a g e

Learn C# - do it yourself way...

2008

using System; namespace payroll { public class Employee { //Body of the class } public class Salary : Employee { //Body of the class } public class Hourly : Employee { //Body of the class } }

The Employee, Salary and Hourly classes are now all in the same namespace.

8.7 Compartments created by Namespaces Namespace creates a compartment for all classes. If we put a class inside a namespace/compartment, the namespace becomes a part of the name of the class. Say for example, when we put Employee class inside payroll package, the name of the class now becomes payroll.Employee

The Employee class can now no longer be accessed by the name Employee. To refer to the Employee class, we must everytime call it payroll.Employee. Similarly, we must refer to Salary and Hourly classes as payroll.Salary and payroll.Hourly. Suppose you have a written a Vehicle class, and I too have written a Vehicle class. Kate wants to use both the Vehicle classes in their program. So, I put my Vehicle class in a compartment quasar, and you put your vehicle class in another compartment student. Kate will now refer to both the Vehicle classes as,

33 | P a g e

Learn C# - do it yourself way...

2008

quasar.Vehicle student.Vehicle To create objects of these classes, Kate would write quasar.Vehicle v1 = new quasar.Vehicle(); student.Vehicle v2 = new student.Vehicle(); Here, v1 refers to a quasar.Vehicle object, whereas v2 refers to student.Vehicle object. Thus, from the above example, you can deduce the fact that, namespaces help you to avoid naming conflicts by creating compartments. Note – Classes within the same namespace/compartment do not need to use the . convention while referring to each other. Thus, if write another class Test inside payroll namespace, we can call Employee, Salary and Hourly classes directly without using any special naming convention. Classes in the same compartment find each other without any special syntax.

8.8 using Keyword If a class wants to refer to another class within the same namespace/compartment, the namespace need not be used. We could refer to it directly. However, outside the compartment, we must use the fully qualified name of the class. Consider the following program : using System; namespace payroll { public class Employee { public string name; public string address; public int SSN; public int number;

34 | P a g e

Learn C# - do it yourself way...

2008

public Employee(string n, string a, int ssn, int num) { name = n; address = a; SSN = ssn; number = num; } } }

There is another class Boss which would like create Employee objects. using System;

class Boss { public static void Main() { payroll.Employee e = new payroll.Employee("Jane Smith", "111 Oak drive", 999001111, 1); } }

Writing payroll.Employee everytime could become tedious or cumbersome. To make things easy, we can use the using keyword. using keyword specifies in which namespace to look for the given classes. If you write using payroll, then we can refer to the Employee class, simply as Employee without using the fully-qualified name. using System; using payroll; class Boss { public static void Main() { Employee e = new Employee("Jane Smith", "111 Oak street", 999001111, 1); } }

Note that, using keyword is just a convenience statement. During compilation, the C# compiler will replace all instances of Employee class with payroll.Employee(Fully-Qualified name).

35 | P a g e

2008

Learn C# - do it yourself way...

8.9 Skeleton of a C# Executable Assembly MyProgram.exe

File1.cs

File2.cs

File3.cs

namespace A { .... }

namespace B { .... }

namespace C { .... }

class A { //Body of class }

class B { //Body of class }

class C { //Body of class }

8.10 Interfaces An interface is a collection of abstract members. A class implements an interface thereby inheriting the abstract methods of the interface. All the methods of an interface need to be defined in the class. Syntactically, an interface is defined using the C# interface keyword. public interface IName { void method1(); void method2(); }

In the .NET Framework, interfaces have a special naming convention. All interface names begin with a capital I. Interfaces can have properties as well. Difference from a class – 1. You cannot create an object of an interface. 36 | P a g e

Learn C# - do it yourself way...

2008

2. An interface does not contain any constructor. 3. All methods in an interface are abstract. 4. An interface can inherit multiple interfaces. An interface is like a contract – a promise that a class will implement a specific set of functionalities. When a class implements an interface, it signs a contract, a treaty and promises to provide implementation/definition of all methods or functions declared in the interface. C# code can question an object to determine whether it supports an interface. Interrogating an object is basically asking the question,”Do you support this interface?”. If the object answers, “Yes, I do!”, than you can call the methods defined in the interface on the object.

8.11 Exposing Methods through an Interface public class Employee : Payable,EmployeeInfo { string name, address; double weeklyPay; public Employee(string n, string a) { name = n; address = a; } public string GetName() { return name; } public string GetAddress() { return address; } public void SetName(string n) { name = n; } public void SetAddress(string a) { address = a; }

37 | P a g e

Learn C# - do it yourself way...

2008

public double GetWeeklyPay() { return weeklyPay; } public void computePay(int hoursWorked) { weeklyPay = hoursWorked * 6.50; Console.WriteLine("Weekly Pay for : " + name + " is Rs." + weeklyPay); } public void MailCheck() { Console.WriteLine("Mailing check to : " + name + " at " + address); } }

An organisation has many employees working in it. The Employee of an organisation is represented by an Employee class object. Associated with every Employee is his personal information such as his name, address. Every employee also has a salary and other payrelated information. The organisation has two departments – Payroll and Human Resource. Different parts in an organisation have different data needs. For example, the payroll department handles the payroll needs, but it does not need access to or change the personal information of an employee. On the same lines, the Human Resource department manages the general information about Employees, but it does not need access to the Employee’s pay details. How do we realise these business rules in C#? The answer lies in the innovative use of Interfaces. Although we cannot instantiate(create an object) an interface, we can do the following –

38 | P a g e

Learn C# - do it yourself way...

2008

interface MyInterface { void a(); void b(); } class MyClass : MyInterface { void a(){ Console.WriteLine(“Inside A”); } void b(){ Console.WriteLine(“Inside B”); } void c(){ Console.WriteLine(“Inside C”); } } Declare an interface reference MyInterface iref; Assign to it, an object of a class that implements this interface iref = new MyClass(); Using iref, we can call the methods a() and b(), but we cannot call the method c(). Thus, we can call only those methods on the object, which are declared and exposed by the interface. Thus, by assigning an object to an interface reference, we can restrict access to the methods that can be called on the object. In the Employee class example, we could take two different interfaces payable and EmployeeInfo for the Payroll and HR departments. The payable interface exposes and provides an interface to only those 39 | P a g e

Learn C# - do it yourself way...

2008

methods that are needed by the payroll department. Through the EmployeeInfo interface, the HR department can only access those methods which help it to manage Employee Information like name and address. public interface Payable { void computePay(int hoursWorked); void MailCheck(); double GetWeeklyPay(); } public interface EmployeeInfo { string GetName(); void SetName(string name); string GetAddress(); void SetAddress(string address); }

To represent the Payroll and Human Resource Departments, we write the Payroll and HumanResource classes. public class Payroll { public void PayEmployee(Payable p) { p.computePay(10); p.MailCheck(); } } public class HumanResource { public string GetInfo(EmployeeInfo e) { return e.GetName() + " " + e.GetAddress(); } public void ChangeName(EmployeeInfo e, string n) { Console.WriteLine("Changing Name for : " + e.GetName()); e.SetName(n); Console.WriteLine("New name is : " + e.GetName()); } public void UpdateAddress(EmployeeInfo e, string a) { Console.WriteLine("Changing address for : " + e.GetName()); e.SetAddress(a); Console.WriteLine("New address is :" + e.GetAddress()); } }

40 | P a g e

Learn C# - do it yourself way...

Now, we write the Main Program, where we take a fictious Employee, and the payroll and HR departments. class EmployeeManagement { public static void Main() { Employee e = new Employee("George Washington","Mt. Vernon"); Payroll payroll = new Payroll(); HumanResource hr = new HumanResource(); payroll.PayEmployee(e); hr.GetInfo(e); hr.ChangeName(e, "Bill Gates"); hr.UpdateAddress(e, "Redmond VA"); } }

41 | P a g e

2008

Learn C# - do it yourself way...

2008

9 File Reading and Writing – I/O 9.1 Concept of Streams Many applications which you would write, need to store data and retrieve data from a file. For example, in an Employee Management System, you would like the Employee details to be stored permanently in a file on the disk. You would also want to read Employee data from the file on disk. You might also need to update an employee’s information in a file, say for example his salary is raised, or his marital status changes from single to Married. Reading or writing data to files is a functionality which you require often while making your application. Thus, it is important that we study, how we can read or write to files. However, an application cannot directly read or write data to a file. To connect an application to a file, we must use a virtual pipe. Just as a real pipe carries water, the virtual pipe carries bytes of data to and fro between the application and the file. This virtual pipe in C# is called a Stream. If the stream connects to a file, its called FileStream.

42 | P a g e

Learn C# - do it yourself way...

2008

To create a FileStream in C#, we must create an object of the FileStream class. FileStream fs = new FileStream(,,,,) The FileStream constructor accepts the FileName as the first argument. This must be the absolute file path. Next, the file mode must be specified. The file mode tells the C# system, which mode you would like to operate the file in. The FileMode enum looks like this – enum FileMode{ CreateNew, Create, Open, OpenOrCreate, Truncate, Append }

43 | P a g e