This page

intentionally left
blank
Copyright © 2009, New Age International (P) Ltd., Publishers
Published by New Age International (P) Ltd., Publishers

All rights reserved.

No part of this ebook may be reproduced in any form, by photostat, microfilm,
xerography, or any other means, or incorporated into any information retrieval
system, electronic or mechanical, without the written permission of the publisher.
All inquiries should be emailed to [email protected]

ISBN (13) : 978-81-224-2710-3

PUBLISHING FOR ONE WORLD

NEW AGE INTERNATIONAL (P) LIMITED, PUBLISHERS
4835/24, Ansari Road, Daryaganj, New Delhi - 110002
Visit us at www.newagepublishers.com
C++ is an excellent programming language both for procedural as well as object oriented
programming and lends support to more diverse applications compared to any other computer
programming language. This one reason that C++ now is used in more than 80% of all the
softwares that is being produced these days. Naturally, those with sound knowledge of C++
have access to many job-opportunities. However, many students feel that programming with
C++ is a difficult task. In this book, the presentation of the subject has been designed keeping
in view the difficulties of the students. Almost every topic is followed by illustrative examples,
so that the readers may easily grasp the content.
In programming, one learns much more quickly through actual live programs than by
simply reading volumes about how to program. Every chapter of the book contains illustrative
examples in the form of well designed live programs which the reader may use for learning
as well as for his/her applications. The programs are designed to bring out the salient features
of the underlying principle or coding technique. The book contains more than 450 live
programmes. This feature of the book is highly useful for those doing self study of C++
language.
The book also contains an extensive coverage of STL and other recent additions to C++
Standard Library. STL contains function templates and class templates which are meant to
be reused by professionals in their application programmes. In fact, the reuse of well
established software is at the heart of object oriented programming. The book contains a
large number of live programs for illustrating the applications of the STL functions and
algorithms. Thus the book is highly useful to professionals as a ready reference.
The first few chapters of the book deal extensively with the procedural programming.
This style of programming is still used extensively for small length programs. Also, a good
grounding in procedural programming prepares the students for more difficult topics pertaining
to classes, operator overloading, inheritance, object oriented programming (OOP), template
functions and template classes as well as container classes such as vector, list, deque, sets,
multisets, maps, multimaps, stacks, queues, etc. All these topics are dealt in sufficient detail
as well as a large number of illustrative live programs are included. This part is highly useful
to students as well as software professionals. These templates are meant to be reused in
professional application software.
❖ vi ❖ Programming with C++

Almost all the topics generally used in C++ programming are covered in the book. The
contents of the book have already been class tested. From the responses of students, the authors
feel confident that the book should be useful for students pursuing science and engineering,
BCA, MCA and those appearing for examinations of Department of Electronics Society.
Authors are also confident that the book will prove its worth to software professionals as
well as to those pursuing self-study.

B.L. Juneja
Anita Seth
 Contents ❖ vii❖

The authors would like to record their gratitude to the Director of IIT Delhi and Director
of Institute of Engineering and Technology, Indore, for providing general facilities for carrying
out the writing of this book. Authors are grateful to the Head of Department of Electronics
and Telecommunication, IET Indore, and Prof. J.P. Subrahmanyam, Head of Department
of M.E., IIT Delhi for encouragement and inspiration.
The authors are particularly grateful to Professor Arun Kanda and Professor S.G.
Deshmukh – the previous Head and Associate Head of Mechanical Engineering Department,
IIT Delhi, for providing facilities to Dr. B.L. Juneja to teach the subject to the graduate
students of Mechanical Engineering Department. This has given him the opportunity to test
the material in class environment. We are highly grateful to Prof. P.V.M. Rao of Mechanical
Department, IIT Delhi for allowing use of computer laboratory and helping with necessary
software.
Authors are also grateful to Professor G. Singh of Maryland University, USA for his
valuable advice, encouragement and for providing the syllabi of the courses on C++ being
taught at American Universities.
Authors are grateful to Dr. Nitin Seth of Mechanical Department, IET Indore, presently
working as Professor, Indian Institute of Foreign Trade, New Delhi for his valuable suggestions
and general help in presentation of material.
For accomplishing a work like writing of this book, it requires, besides the efforts of
the authors, the advice and help of the persons closely associated with the authors. The authors
are indeed grateful to Mr. Rajiv Kapoor, Mr. Bharat Kalra and all the colleagues for their
valuable suggestions and encouragement.
The last but not the least, authors are indeed grateful to Mrs. Asha Seth and Mrs. Pushpa
Juneja for providing the moral support and help so that the work could be completed.

B.L. Juneja
Anita Seth
 This page
intentionally left
blank
Preface v
Acknowledgements vii

Chapter 1: Introduction to C++ 1

1.1 Computer and Computer Languages 1
1.2 A Brief History of C++ 4
1.3 Major Additions from C to C++ 5
1.4 Some Advantages of C++ Over C 6
1.5 Range of Applications of C++ 7
1.6 Compilers for Programming with C++ 7
1.7 The C++ Standard Library 8
1.8 Program Development in C++ 9
1.9 Programming Techniques 10
1.10 Object Oriented Programming 12
1.11 Operator Overloading 15
1.12 Inheritance 16
1.13 Polymorphism 16
1.14 Number Systems 17
1.15 Bits and Bytes 20
1.16 Computer Performance 23
Chapter 2: Structure of a C++ Program 27
2.1 Introduction 27
2.2 Components of a Simple C++ Program 28
2.3 Escape Sequences 30
2.4 Variable Declaration and Memory Allocation 34
2.5 Namespaces 36
2.6 User Interactive Programs 38
2.7 Formatting the Output 39
2.8 Function cin.getline() v/s cin 45
❖ x ❖ Programming with C++

Chapter 3: Fundamental Data Types in C++ 53

3.1 Fundamental Data Types 53
3.2 Declaration of a Variable 57
3.3 Choosing an Identifier or Name for a Variable 58
3.4 Keywords 59
3.5 Size of Fundamental Data Types 61
3.6 Scope of Variables 66
3.7 Type Casting 68
3.8 The typedef 71
3.9 The typeid () Operator 72
3.10 Arithmetic Operations on Variables 73
3.11 Function swap() 73
Chapter 4: Operators 77
4.1 Introduction 77
4.2 Assignment Operator 78
4.3 Arithmetic Operators 82
4.4 Composite Assignment Operators 85
4.5 Increment and Decrement Operators 86
4.6 Relational Operators 89
4.7 Boolean Operators 90
4.8 Bitwise Operators 91
4.9 Precedence of Operators 95
Chapter 5: Selection Statements 99
5.1 Introduction 99
5.2 Conditional Expressions 100
5.3 The if Expression 100
5.4 The if … else Statement 102
5.5 Conditional Selection Operator ( ? : ) 104
5.6 The if … else Chains 105
5.7 Selection Expressions with Logic Operators 107
5.8 The switch Statement 109
Chapter 6: Iteration 117
6.1 Introduction 117
6.2 The while Statement 117
6.3 The Nested while Statements 119
6.4 Compound Conditions in a while (expression) 121
6.5 The do….while Loop 123
6.6 Endless while Loops 126
 Contents ❖ xi❖

6.7 The for Loop 127

6.8 Compound Conditions in for (expression) 130
6.9 Nested for Loops 131
6.10 Generation of Random Numbers 132
6.11 The goto Statement 136
6.12 The continue Statement 138
6.13 Input with a Sentinel 138
Chapter 7: Functions 144
7.1 Introduction 144
7.2 User Defined Functions 145
7.3 Function Prototype and Return Statement 147
7.4 Integration of a Function 152
7.5 Functions with Empty Parameter List 153
7.6 Function Overloading 153
7.7 The inline Functions 154
7.8 Use of #define for Macros 155
7.9 C++ Standard Library Functions 157
7.10 Passing Arguments by Value and by Reference 160
7.11 Recursive Functions 163
Chapter 8: Arrays 167
8.1 Declaration of an Array 167
8.2 Accessing Elements of an Array 169
8.3 Input/Output of an Array 170
8.4 Searching a Value in an Array 177
8.5 Address of an Array 179
8.6 Arithmetic Operations on Array Elements 180
8.7 Sorting of Arrays 183
8.8 Finding the Maximum/Minimum Value in an Array 186
8.9 Passing an Array to a Function 187
8.10 Two Dimensional Arrays 189
8.11 Two Dimensional Arrays and Matrices 191
8.12 Three Dimensional Arrays (Arrays of Matrices) 194
Chapter 9: Pointers 197
9.1 Introduction 197
9.2 Declaration of Pointers 197
9.3 Processing Data by Using Pointers 202
9.4 Pointer to Pointer 203
9.5 Pointers and Arrays 205
❖ xii ❖ Programming with C++

9.6 Array of Pointers to Arrays 208

9.7 Pointers to Multi-dimensional Arrays 210
9.8 Pointers to Functions 212
9.9 Array of Pointers to Functions 214
9.10 Pointer to Functions as Parameter of Another Function 215
9.11 The new and delete 216
9.12 References 218
9.13 Passing Arguments by Value and by Reference 221
9.14 Passing Arguments Through Pointers 224
9.15 Pointer Arithmetic 226
9.16 Void Pointers 227
9.17 Summary of Pointer Declarations 228
Chapter 10: C-Strings 231
10.1 Declaration of a C-string 231
10.2 Input/Output of C-strings 234
10.3 Standard Functions for Manipulating String Elements 239
10.4 Conversion of C-string Characters Into Other Types 241
10.5 Arrays of C-strings 244
10.6 Standard Functions for Handling C-string Characters 245
10.7 Standard Functions for Manipulation of C-strings 247
10.8 Memory Functions for C-strings 255
Chapter 11: Classes and Objects-1 261
11.1 Introduction 261
11.2 Declaration of Class and Class Objects 262
11.3 Access Specifiers – private, protected and public 264
11.4 Defining a Member Function Outside the Class 266
11.5 Initializing Private Data Members 267
11.6 Class with an Array as Data Member 270
11.7 Class with an Array of Strings as Data Member 271
11.8 Class Constructor and Destructor Functions 277
11.9 Types of Constructors 278
11.10 Accessing Private Function Members of a Class 281
11.11 Local Classes 283
11.12 Structures 284
Chapter 12: Classes and Objects-2 291
12.1 Friend Function to a Class 291
12.2 Friend Classes 294
12.3 Pointer to a Class 296
 Contents ❖ xiii❖

12.4 Pointers to Objects of a Class 299

12.5 Pointers to Function Members of a Class 300
12.6 Pointer to Data Member of a Class 301
12.7 Accessing Private Data of an Object Through Pointers 302
12.8 The this Pointer 305
12.9 Static Data Members of a Class 307
12.10 Static Function Member of a Class 308
12.11 Dynamic Memory Management for Class Objects 310
12.12 A Matrix Class 312
12.13 Linked Lists 314
12.14 Nested Classes 320
Chapter 13: Operator Overloading 322
13.1 Introduction 322
13.2 Operators that may be Overloaded 324
13.3 Operator Overloading Functions 324
13.4 Addition of Complex Numbers 328
13.5 Overloading of += and –= Operators 329
13.6 Overloading of Insertion (<<), Extraction ( >>) and /= Operators 330
13.7 Overloading of Increment and Decrement Operators (++ and – –) 331
13.8 Dot Product of Vectors 333
13.9 Overloading of Equality Operator (==) 334
13.10 Overloading of Index Operator [] 335
Chapter 14: Inheritance 339
14.1 Introduction 339
14.2 Forms of Inheritances 340
14.3 Single Public Inheritance 342
14.4 Single Protected Inheritance 346
14.5 Single Private Inheritance 348
14.6 Multiple Inheritance 351
14.7 Multilevel Inheritance 352
14.8 Constructors and Destructors in Inheritance 355
14.9 Containment and Inheritance 361
14.10 Overloaded Operator Functions and Inheritance 366
Chapter 15: Virtual Functions and Polymorphism 369
15.1 Introduction 369
15.2 Virtual Functions 369
15.3 Arrays of Base Class Pointers 376
15.4 Pure Virtual Functions and Abstract Class 377
❖ xiv ❖ Programming with C++

15.5 Virtual Destructors 379

15.6 Virtual Base Class 383
15.7 Run-time Type Information (RTTI) 385
15.8 New Casting Operators in C++ 388
Chapter 16: Templates 396
16.1 Introduction 396
16.2 Function Templates 396
16.3 Function Template with Array as a Parameter 400
16.4 Function Templates with Multiple Type Arguments 404
16.5 Overloading of Template Functions 406
16.6 Class Templates 407
16.7 Friend Function Template to Class Template 410
16.8 Friend Class Template to a Class Template 413
16.9 Template Class for Complex Variables 415
Chapter 17: C++ Strings 417
17.1 Introduction 417
17.2 Construction of C++ Strings 417
17.3 C++ String Class Functions 419
17.4 Applications of Some C++ String Class Functions 421
17.5 String Class Operators 424
17.6 Array of C++ Strings 428
Chapter 18: Exception Handling 431
18.1 Introduction 431
18.2 The try, throw and catch 432
18.3 Catch all Types of Exceptions 438
18.4 Exception Handling Function 440
18.5 Exception Specification 441
18.6 Rethrow an Exception 444
18.7 C++ Standard Library Exception Classes 446
18.8 Function terminate() and set_terminate() 448
18.9 Function unexpected() and set_unexpected() 449
18.10 The auto_ptr Class 450
Chapter 19: Input/Output Streams and Working with Files 453
19.1 Introduction 453
19.2 I/O Streams for Console Operations 454
19.3 Predefined Standard I/O Streams 455
19.4 Functions of <istream> and <ostream> 458
19.5 Formatted I/O Operations with Manipulators 464
 Contents ❖ xv❖

19.6 Formatting by Setting Flags and Bit Fields 466

19.7 Introduction to Files 470
19.8 File Stream Classes 472
19.9 File Input/Output Streams 473
19.10 Functions is_open(), get() and put() for Files 478
19.11 The Function open() and File Open Modes 479
19.12 File Pointers 481
19.13 Binary Files and ASCII Character Codes 484
19.14 Functions write() and Read() for File Operations 485
19.15 File Operations for Class Objects 486
19.16 Random Access Files 488
19.17 Error Handling Functions 488
Chapter 20: Namespaces and Preprocessor Directives 493
20.1 Introduction to Namespaces 493
20.2 Application of Namespaces 493
20.3 Directives using and using namespace 495
20.4 Namespace Aliases 496
20.5 Extension of Namespaces 498
20.6 Nesting of Namespaces 499
20.7 The namespace std 500
20.8 Preprocessor Directives 501
20.9 Conditional Preprocessor Directives 504
20.10 Predefined Macros 507
Chapter 21: Standard Template Library 509
21.1 Introduction 509
21.2 The Containers 510
21.3 Iterators 511
21.4 Sequence Containers 513
21.5 Associative Containers 518
21.6 Container Adapters: Stack, Queue and Priority Queue 519
21.7 Function Objects/Predicates 526
21.8 Predefined Predicates in C++ 528
21.9 Binder Functions 530
Chapter 22: Sequence Containers–vector, list and deque 532
22.1 Introduction 532
22.2 Vector Class 532
22.3 Functions of Vector Class 534
22.4 Definition and Application of Iterators 535
❖ xvi ❖ Programming with C++

22.5 Operators Supported by Vector Class 537

22.6 Application of Functions of Vector Class 537
22.7 Functions and Operators Supported by List Class 546
22.8 Application of Some Functions of List Class 548
22.9 Functions and Operators Supported by Deque 553
22.10 Application of Some Functions Supported by Deque 554
Chapter 23: Associative Containers–set, multiset, map and multimap 559
23.1 Introduction 559
23.2 Functions Supported by Associative Containers 560
23.3 The Sets 561
23.4 The Multisets 566
23.5 The Maps 567
23.6 The Multimaps 574
Chapter 24: Bit Sets 578
24.1 Introduction 578
24.2 Construction of Bit Sets 579
24.3 Bit Set Operators 581
24.4 Bitset Class Functions 584
Chapter 25: Algorithms 589
25.1 Introduction 589
25.2 Categories of Algorithms and Brief Descriptions 590
25.3 Illustrative Applications of Some Algorithms 594
REFERENCES 625
APPENDIX–A ASCII CHARACTER CODE SET 627
APPENDIX–B C++ KEYWORDS 631
APPENDIX–C C++ OPERATORS 634
APPENDIX–D COMMONLY USED HEADER FILES IN C++ 636
APPENDIX–E GETTING STARTED 638
SUBJECT INDEX 647
CHAPTER

++

1.1 COMPUTER AND COMPUTER LANGUAGES

Computer has become an essential element of our modern society. Work in offices, factories,
selling outlets, research and development laboratories, teaching, telecommunication, sports, and
even home entertainment are all being controlled by computers. Not only that, most of our
future developments are closely linked with the developments in computer systems which involve
development in hardware as well as software. This is reflected in the rapid developments in
computer hardware that we see today. Super computers that are capable of doing trillions of
instructions per second is the talk of the day. The personal computers whether desktops or laptops
have capabilities and speeds much better and prices much less than the mainframe computers of
just a decade ago. Only machines are not enough. Robust and efficient software is needed to make
these machines to perform to the fullest extent. Computer languages are the bases of efficient
software.
A typical computer hardware (Fig. 1.1) consists of one or more microprocessor chips
connected to the supporting chips and other electronic circuits. These chips include cache and
primary memory chips usually called RAM (Random Access Memory). Besides, the
microprocessor is also connected to bulk storage memory devices such as floppy disc, CD ROM
(Compact Disc Read only Memory), hard disc and magnetic tapes etc. The microprocessor chip
which is the main chip is also called CPU (Central Processing Unit). Its internal structure consists
of circuits which form a number of registers (typical number is 16), an arithmetic unit, a logic
unit and a control unit. It has large number of terminals for connection to external devices (for
instance a typical CPU has as many as 478 terminals). Out of these a group of terminals is for
data input and data output, another group is for memory addresses and still another group for
control purposes. Besides, several terminals are for interrupt service. An interrupt is like gate
crashing. Suppose some program is being processed in the computer and you interrupt that program
and start your own program. The computer first saves the ongoing program status, takes up your
item, and after finishing work on it, it resumes the ongoing program. Every CPU supports a set
of instructions designed by its manufacturer. We need not go into details of the complex circuitry,
however, some of the names mentioned above are often used in software and a programming
student should be familiar with them. A beginner in programming will most probably be using a
desktop personal computer (PC). A typical PC cabinet is illustrated in Fig. 1.2. The input/output
devices that a reader would in general come across are shown in Fig. 1.3.
❖ 2 ❖ Programming with C++

Fig. 1.1: Components of a computer system. Bus lines carry electrical pulses
from one device to another. Arrows show the data flow

Fig. 1.2: A typical desktop computer cabinet. The various ports

may be placed differently on different models

Output port

Computer circuits
A
code

Input
port
Code output
Voltages
5 000005 Display of output
applied
Monitor - Default output device
1000001 for C++
Code

Input from Keyboard - Default input device

for C++
Fig. 1.3: A PC (personal computer) and standard input/output devices
 Introduction to C++ ❖ 3❖

MACHINE LANGUAGE
As already mentioned, a computer is a collection of electronic circuits driven by application of
high voltage (typical value is 5 volts) at some points and low voltage (typical value is 0) at other
points in the circuit. We may designate (code) high voltage as 1 and low voltage as 0. Thus the
basic code for computer instructions consists of sequences of 1s and 0s such as 00111110. The
values of variables, which are numbers, can also be represented in binary numbers (discussed later
in the chapter) which are also sequences of zeros and ones. Thus a program which comprises
instructions and data may be written as a series of sequences of zeros and ones. There are different
instruction formats (combinations of opcodes, codes and data) so that the processor can make
out which sequence of 0s and 1s is instruction and which is data. The first number in an
instruction is opcode followed by data. Some instructions may not have any data to act upon.
The complete set of these instructions for a CPU is called machine language code or simply
machine language. This is the language which machine understands.
A program written in machine language will consist of a large number of sequences of 0s
and 1s. It is very difficult to write such programs and debug them in case there is an error. Besides,
the instruction codes for processors made by different manufacturers are different for the same
process such as addition or storage of a value etc., therefore, the program written in one machine
language for one microprocessor will only work on computers having similar microprocessor but
will not work on computers based on other microprocessors. That is to say, there is no portability
of these programs. Moreover, it is very cumbersome to write and debug programs in binary codes
or to convert a program made for one make of CPU to another make of CPU. These two factors
have provided impetus for development of high level programming languages which should
be easy to understand, debug and be portable.
Assembly language is the next development. In this the instructional codes have been
converted to more readable forms, for example, for one processor instruction such as MOV AX
116 means move the immediate data that is 116 to destination register AX. The word MOV
is short form of MOVE. Assembly language instructions such as MOV, ADD which stands for
addition, SUB which stands for subtraction etc. are called mnemonics. A program written in
assembly language needs a software called assembler which converts the program consisting of
mnemonics and data into machine language. In this case also the programs are not portable.
Nevertheless, the programs written in assembly language are easier to read as well as easier to
debug than those written in machine language.
In the development of a high level language there has always been an endeavour to make the
language as near to the language that we speak as possible. It is worthwhile to note that the
programs written in machine language are the most efficient in execution, because there is nothing
between the code and the machine as is the case with high level languages. All high level computer
languages need a compiler or interpreter to convert the program instructions and data into
machine language for its execution, because, the machine understands only its own language, i.e.
sequences of 0s and 1s.
❖ 4 ❖ Programming with C++

C LANGUAGE
BCPL was one of the earliest high level computer language developed in 1967 by Martin Richards
for writing operating system which is a software needed to control various processes in computer
applications. Later the language B was developed by Ken Thompson who added many new
features in the language.
C was developed mainly from B by Dennis Ritche. Soon C became popular with
programmers and it remained so for several decades. It is still used in several applications. It is a
structured high level language. Programs written in C are easy to write and debug as well as the
programs are portable. The code is nearest to the assembly language and thus is quite efficient.
However, the code written in C language needs a compiler to convert the different instructions
and data into machine language.
Because of popularity of C many companies produced compilers for C. A compiler is a
software which converts the program into machine language. Because of the widespread use of
C, it became necessary to standardize the syntax and functions of the language so that the
programs could run with compilers developed by different organizations and on different plat-
forms. The American National Standard Institute (ANSI) prepared a standard document for C
as a national standard which later with some modifications was also adopted by ISO
(International Organization for Standardization) as an international standard. The first standard
C document was published in 1990. It is known as ISO/IEC 9899-1990, Programming
Language – C. The C language according to this document is also called C-90.
With further development of C, it became necessary to revise the earlier standard. The revised
standard was ratified by ISO in 1999. The document is known as ISO/IEC 9899-1999,
Programming language – C. The language according to this standard is known as C - 99 and
it is in vogue at present.
In order to meet the variety of demands from the real world problems, several computer
languages have been developed over the same period. These languages are more tuned to the
typical applications that these languages are put to. The number of these languages is quite large
and it is difficult to give even brief discussions of all these languages. Many of these could not
gain any popularity and are dead. However, some of these became popular and have been in use
for a number of years. The popular ones are Basic and Visual Basic, Cobol, Fortran, Logo, Prolog,
Pascal, Lisp, Ada, Simula, Smalltalk, etc. After the development of C++, another language Java
has been developed. Java is more user friendly. However, C++ is a very versatile language and finds
applications in many diverse fields and hence is very popular.

1.2 A BRIEF HISTORY OF C++

In eighties an average large C program was of the order of a few thousands of lines of code. With
really large programs the difficulties with C language became more apparent and researchers were
busy finding better options. C++ is a redevelopment of C carried out by Bjarne Stroustrup when
he was working with AT&T during 1983-1985. Inspired by the language Smalltalk and Simula
which are object oriented programming (OOP) languages, he added the concept of classes, in
order to impart similar capabilities in the then widely popular C language. Classes are the basis
 Introduction to C++ ❖ 5❖

of OOP. Initially the new language was called C with classes. The name C++ was coined later
by Rick Mascitti in 1983. The symbol (++) is an operator in C which increases the value of an
integer variable by 1. The name C++ aptly shows that C++ is oneup on C. The C++ is also often
called as better C.
The C++ programming language became popular soon after it was officially launched. As a
result, scores of programmers got involved, who have not only tested the codes and recognised
the pitfalls, but have also added reusable program modules to C++ Standard Library which is a
large collection of well tested and reusable programs which facilitate the writing and execution
of diverse applications. It became necessary to standardise C++ as well. So an ISO committee
was formed in 1995 and C++ standard was ratified by ISO in 1998. The document is officially
known as ISO/IES 14882-1998, Programming Language – C++. The language according to
this standard is briefly referred to as C++ 98. Later a revise version of the standard with minor
corrections and clarifications was published in 2003. This document is known as ISO/IEC
14882::2003. This is the standard being followed at present. This is also available in the book
form with name “The C++ Standard” and published by John Wiley Ltd.
In fact, there has been continuous development in the language. Addition of inheritance of
classes, operator overloading and polymorphism has enhanced the application of the language.
The addition of template functions and template classes has given rise to generic programming.
A remarkable enhancement to the language is the addition of Standard Template Library as a part
of C++ Standard Library. These programs can be used straightway by the programmers to create
new applications as well as new programs. Addition of namespace has added additional flexibility
in combining programs written by different programmers without running the risk of ambiguity
of same names being used for different variables in different sections of the program. These
developments in C++ have enhanced the capabilities of C++ enormously. A very large program
may be broken into small modules and classes, which may be developed by different programmers
and tested for their performance before combining them into one large program. These
developments have given extensive capability to C++ to have reusable program modules and
classes which comprise the C++ Standard Library.

1.3 MAJOR ADDITIONS FROM C TO C++

There has been continuous effort for upgrading C++ by way of adding new algorithms,
applications and other additions such as namespace, etc. It is really difficult to make a detailed
list of all these items, however, below only major developments are listed which are responsible
for major advantages of using C++ over C.
(1) Classes which provide a mechanism for data abstraction, encapsulation of data and
functions, information hiding and object oriented programming.
(2) Operator overloading.
(3) Inheritance.
(4) Polymorphism.
(5) Addition of namespaces.
❖ 6 ❖ Programming with C++

(6) Template functions and template classes which lead to generic programming.
(7) Addition of STL ( Standard Template Library).
(8) Many new keywords.

1.4 SOME ADVANTAGES OF C++ OVER C

C++ allows procedural programming as well as object oriented programming and generic
programming. In seventies and eighties generally procedural programming were used for short
as well for long programs which comprised a few thousands to hundred thousands lines of
code. The present day big programs comprise millions of lines of code. In procedural
programming, there are many problems in production of really large programs. Debugging is
one major problem. Even after successful implementation the maintenance of the software is
not easy. It is difficult for a new programmer to comprehend the different sections of the
program.
The inclusion of classes has added the advantage of object oriented programming to C++,
though, it is also an efficient language in procedural programming. The inheritance of classes in
C++ has also made easy to reuse the existing class programs. If you have made a class program,
verified and tested and later you wish to add new things into it, you can do it without modifying
the tested and verified program. You simply make another class program which inherits the
existing class and adds the new features into it. The original program remains intact and is used
in the new program without any modification. Later on you can make still another class which
inherits any one or both of the already developed classes and adds new features to the whole
program. This allows a continuous addition of new features in an already running program
without disturbing the existing classes.
Operator overloading gives the facility for applying the same operators that we use for
fundamental types to the class objects as well. Thus the operations on vectors, complex numbers,
strings and other class objects are made easier. The operator overloading gives a large extension
to C++.
The addition of inheritance and polymorphism in C++ has extended the scope of use of
existing software. A big program may be designed to comprise convenient reusable components
or modules comprising different classes. It is possible to extend or modify the components as
explained above in case of classes. The various sub-programs may be re-assembled in different
ways to make new programs. Debugging of a big program becomes easy because each class and
module can be tested and verified independently of the whole program. Another major advantage
is that a number of programmers or teams of programmers may be involved in developing
different modules of a big program which is difficult in procedural programming. In C++ there
is another advantage that these modules may be seamlessly combined without a clash of names
of variables and functions by using namespaces. Also the syntax in C++ is easier than that in C.
It is because of all these advantages that C++ is used in more than 80% of all the software that
is being created these days.
Though C++ is often called as superset of C, however, there are differences as well. Many
students ask this question. Is it necessary to learn C for learning C++? The answer is no.
 Introduction to C++ ❖ 7❖

For learning C++ it is not necessary to first learn C. If you already know C it will certainly help,
because, the concepts in procedural programming are similar. If you do not know C there is
nothing to worry because C++ is best learned on its own and you would be avoiding confusion
because of some differences in syntax in the two languages.

1.5 RANGE OF APPLICATIONS OF C++

C++ has now emerged as one of the most advanced and most popular computer language
developed so far and has highest diversity in its applications. A comprehensive list of applications
has been compiled by Bjarne Stroustrup, the originator of C++ himself and is available on internet
at the following address.
http://www.research.att.com/~bs/applications.html
The list includes the names of companies who have developed many different softwares for
diverse applications and in which C++ has been used. Here we list some of the application fields
in which C++ has been used for producing software. These are listed below. The list is largely
based on the above mentioned reference.

A PARTIAL LIST OF FIELDS OF APPLICATIONS OF C++

(i) Operating Systems and component in windows XP and other OS components.
(ii) Software for e-commerce.
(iii) Customer service software such as electronic ticketing, car and flight reservations, etc.
(iv) Telecommunication services.
(v) Real time control applications.
(vi) 2-D and 3-D animations and games.
(vii) Multiprocessor multimedia software.
(viii) Portable set up box and digital TV software.
(ix) CAD software.
(x) Software for medical instruments and systems.
(xi) Java virtual machine (JVM).
(xii) Telephones services.
(xiii) Mobile phones billing systems and other infrastructural facilities.
(xiv) Irrigation control systems.
(xv) Speech recognition and dictation systems.
The interested reader should consult the above mentioned internet site of Bjarne Stroustrup
for more information.

1.6 COMPILERS FOR PROGRAMMING WITH C++

For execution of programs written in C++ you need to load a compiler software on your
computer. The function of compiler is to compile the program that you have written and present
the code to computer in machine language for execution. Besides this, the compilers in integrated
❖ 8 ❖ Programming with C++

development environment also assist the programmers by pointing out the mistakes in the
program listing. This aspect is very useful for learning C++, because, not only beginners even
experienced programmers would make mistakes in writing the code and if the compiler does not
point it out, the programmers will have to struggle hard to find it out. This an excellent facility
and a beginner should get acquainted with the error pointed out by compiler and the remedial
action. This can best be done by first have a correct program, then introduce different types of
errors one by one and find the response of compiler. A few trials will acquaint a beginner with
the common errors and the remedial actions needed for them.
There are many compilers available in market. Some are free and may be downloaded from
internet. Some are paid but available on trial basis for a specified period while others are on
payment systems. The reader is advised to see following internet sites for obtaining more
information.
(i) An incomplete list of C++ Compilers - by Bjarne Stroustrup at following address.
http://www.research.att.com?~bs?compilers.html
(ii) Google Directory of C++ compilers – at the following address.
http://www.rafb.net/efnet_cpp/compilers
Some of the free compilers for windows are given below.
(i) Borland C++ 5.5
(ii) Dev-C++
(iii) LCC-Win32
(iv) Microsoft Visual C++ Toolkit 2003
For Multi-platform a free compiler is
(i) GCC
For imbedded C++ applications a free compiler is
(i) djgpp for intel 80386 ( and higher)
It is better to start with an integrated development environment (IDE) like MS Visual C++ 6.

1.7 The C++ STANDARD LIBRARY

It is a rich collection of programs on classes, functions and template classes and template
functions which are helpful in program writing and in its execution. Every program requires
applications of several functions of C++ Standard Library. These functions are included in the
program by including the header files which contain these functions (see Chapter 2 for header
files). Standard Template Library (STL) has also been made part of C++ Standard Library. STL
mainly comprises container classes such as vectors, lists, deque, associative containers such as sets,
multisets, maps, multimaps, etc. and near-containers such as C++ strings, etc. Also it has iterator
classes which help in traversing through the containers. These programs can be directly used by
a programmer to create different applications. Besides, there are more than 70 algorithms in STL
which may be used by programmers for creating new applications and programs. All these are,
in general, supported by C++ compilers of good companies.
 Introduction to C++ ❖ 9❖

1.8 PROGRAM DEVELOPMENT IN C++

The start is made by installing one of the compilers described above on the computer. However,
for a beginner it is better to work with an IDE like MS Visual C++ 6, which does text editing,
compiling, linking and makes a binary file of the source code (the program typed by the
programmer) which is loaded on to RAM for execution. A C++ program consists of a sequence
of statements typed by the programmer from top to bottom with the help of text editor. It is
presumed that a beginner will go through the documents provided with the compiler regarding
the instructions for its loading and running. The programmer after typing the program, can direct
for compiling. The programmer may get a list of syntax errors in the typed version of the
program. Some compilers also indicate the location of the error (the line in which error is there)
in the program. The programmer should correct these errors and again direct for compiling.
When the text of the program is error free, the programmer can direct for run or build the
program. The different elements in the process of making, compiling and running a program
are also illustrated in the Fig. 1.4. See Appendix E for starting with Microsoft Visual C++ 6.0.

Fig. 1.4: Sequence of processes in development of a program in C/C++ language.

(Also See Appendix E for more details on starting programs in C++)
❖ 10 ❖ Programming with C++

All these processes are integrated, a programmer has only to type the program, give direction
to compile, correct any error if indicated and then give direction to run the program. The various
processes illustrated in the figure are internal to the compiler.

1.9 PROGRAMMING TECHNIQUES

With the development of programming languages the different programming techniques also
evolved. The distinguishable techniques are listed below.
1. Unstructured or monolithic programming.
2. Procedural Programming
3. Modular Programming
4. Object oriented Programming
For C++ we are mainly concerned with procedural programming and object oriented
programming. However for sake of being able to appreciate the differences, all the techniques
are described briefly below.

UNSTRUCTURED/MONOLITHIC PROGRAMMING
The general structure of a monolithic program consists of global data and statements which
modify the data and finally it contains the output statements. A sample is shown in Fig.1.5.
// Main program
Data
Statement1
Statement2
Statement
——————
Statement1
Statement2
end
Fig. 1.5: Programming in an unstructured programming language
The program is executed from top to bottom, statement by statement. If similar evaluations
are to be carried out at several places in the program, for example, statement1 and statement2
in above illustration, all the statements concerning that evaluation have to be repeated at all the
places where the evaluation is desired. This makes the program very lengthy. Besides, any
modification in the process of the evaluation has to be corrected at so many places in the
program. Such programs are lengthy, difficult to debug and difficult to maintain.

PROCEDURAL PROGRAMMING
These programs are an improvement over the monolithic programs. If a group of statements carry
out similar action at several places in the program, such a group is taken out of the main program
and is placed in a subprogram called subroutine or procedure or function. In the main program
 Introduction to C++ ❖ 11❖

the subroutines or functions are called. When a subroutine is called, the main program is paused
and control shifts to the subroutine till it is finished. Its return value if any is fed to the main
program which is resumed from where it was left. It is illustrated in Fig.1.6.
// Main Program
Global data Procedure1 or Subroutine1
Statement Statement
Statement Statements
——————— —————
Statement
Statement
Statement
Statement Procedure2
————- Statement
Statement Statement
Statement ————

Statement
——————
Statement
end
Fig. 1.6: Procedural programming

A procedural program is more structured. The different procedures or subroutines can be

checked independently of the main program, thus debugging becomes easier. It is also easier to
maintain the program. The drawback is that if the subroutine is small and is called many times
in the program, the overload of calling makes the program inefficient.

Fig. 1.7: Modular programming

❖ 12 ❖ Programming with C++

MODULAR PROGRAMMING
In this type of programming the similar procedures are grouped together into modules. Thus
the whole program may consist of modules and connecting statements. The main program
supplies data and co-ordinates the different modules. It is illustrated in Fig.(1.7).

1.10 OBJECT ORIENTED PROGRAMMING

In this technique an important factor is the data abstraction. The different procedures or
functions are built around the abstract data. Thus the data and functions are encapsulated in a
single structure called ‘class’. The classes are the bases of object oriented programming (OOP).
The classes create new data types besides the fundamental types which are already defined in C++.
In a class, the different functions/procedures are defined according to the data that an object of
the class must have. In the following the concept of classes and objects are explained with
analogies from the physical world.

CLASSES AND OBJECTS

Every object in the real world is unique in itself. However, there are some common characteristics
that it shares with other objects and which give a feeling that there is a group of objects which
are different from others and form a class. For instance, a common characteristic of all fishes is
that they live in water. So fishes form a class of objects that is different from the class of those
who live on ground. Among the fishes there must be many classes or groups such as star fish,
whales, jelly fish, etc., which can be differentiated because of their size, shape, colour, behaviour,
etc. These characteristics give them a separate identity and they can be taken in separate groups
or classes though all of them live in water, which is a common characteristic. Therefore, the
different classes of fish may be derived from a common class, say class Fish. This is called
inheritance. Thus we may make a program to describe the common characteristics of all fishes
while the programs that describe the characteristics of only whales or only star fish form the
classes which are inherited from or are derived from the class Fish.
Let us take another example, that of humans, which form a class which is different from
those of tigers, cats and dogs. But all animals have some characteristics common among
themselves. All belong to a class which we may call class Animals. Among the humans also there
are two distinct classes, that of men and women. So the class Men may be derived from class
Humans which in turn is derived from class Animals. The data and functions which form the
class Animal apply to all the classes downstream that are derived from it, i.e. that of class Men
and class Tigers and class Women. For the class Tiger a particular tiger say the one in the zoo is
an object of this class. Also the authors and the reader of this book are objects or instances of
class Humans. The class Humans would describe the general characteristics of all humans.
Similarly class Men describes the specialised functions which apply only to men. Any particular
man is an instance or an object of the class Men. The same argument applies to many other
objects. For instance, a line is made by points, polygonal shapes are made by lines. Thus we may
make a class Point which deals with points, class Line may be derived from class point and so
on.
 Introduction to C++ ❖ 13❖

DATA ABSTRACTION
Every object of real world has a large amount of data associated with it. The data describes various
characteristics, some of which may be common with other objects. But for a program which is
limited by its purpose and scope some of the data associated with objects of a class may not be
useful for the purpose of the program and hence may not be needed. We have to select the data
most appropriate for the purpose of the program as well as it should be uniformly applicable
to all the objects of the class, for which the program is being prepared. In fact, if we make a
model of the behaviour of the objects of the class, the desired data in the model is the abstract
data.

CLASSES V/S PROCEDURAL PROGRAM

In C++ a class is a sort of blueprint or design in programming that describes the characteristics
of a group of objects which share some common characters. A class comprises data members and
function members. Some of the function members are defined with data members as parameters
while others may serve as an interface for other functions declared private in the class, still others
may be friend functions, etc. When the class is implemented, the functions members operate on
the actual data of an object to bring out characteristics of the object. The actual object data is
substituted for the data members in the functions contained in the class.
One may ask the question, what is the difference between a class and a procedural program
because both involve data and functions? The class program differs in many ways. In case of
procedural program if it comprises the same functions and data as in the class, can certainly bring
out characteristics of one object. For second object, the program is again run with changed data.
However, in case of class, several objects can co-exist each having its own data, while the functions
in class are shared by all the objects. Some function may involve interaction of objects of different
classes.
The class has also the advantage of information hiding. The user need not know how the
program is being implemented. Also the user can not have direct access to object data if the same
is declared private in the class. Thus it enhances the data security. The classes can also be inherited
by another class. The process of inheritance is not limited, it is endless. Thus inheritance makes
it possible to reuse the program in a large number of other programs without any modification.
Still another question remains. Why should we call classes as object oriented programming ?
This question deals with the actual implementation of classes. In a procedural program a user
may call any function and may operate it with data already in the program or with user input
data as planned in the program. In fact, the whole program is before the user who can easily
temper with data members and functions. In case of class, the function call is linked with an
object of the class. The functions of a class can not be called without linking it with the object.
Thus a class program gets implemented only if at least one object is validly defined with its data
on which the functions of the class may operate. A class program is only for its objects. The kind
of data that the objects must possess so that they can become objects of the class is also defined
by the constructor function or another public function in the class declaration. Below is an
example of simple class with name Rectangle. Statements that follow double slash (//) are
comments and are not part of the program. The first 9 lines comprise the class declaration.
❖ 14 ❖ Programming with C++

The main program starts after this. At this stage many readers are not acquainted with some of
the statements and key words written below, nevertheless, one can make out the difference
between this and the procedural programming.

PROGRAM 1.1 – Illustrates a class program.

class Rectangle // declaration of class with name Rectangle

{ // class body starts with {.
private : // access label
int Length ,Width; // private data
public : // access label
void Setsides ( int a, int b){Length= a , Width = b;}
//void Setsides() is public function for accessing private data
int Area (){return Length * Width;}
}; // class declaration ends here.

#include <iostream> //main program starts from here

using namespace std;

int main() // Main function

{
Rectangle Rect1, Rect2; // Rect1 and Rect2 are two objects of
// class Rectangle
Rect1.Setsides(50,40); // Rect1 calls function Setsides().See
// the dot operator (.) between Rect1 and Setsides().
Rect2.Setsides(30,20); // Rect2 calls function Setsides()
cout<<“Area of Rect1 = “<< Rect1.Area() << endl;
//Rect1 calls function Area(). See the dot operator between the
//Rect1 and Area.
cout<<“Area of Rect2 = “ <<Rect2.Area()<<endl;
// Rect2 calls function Area()
return 0 ;
} // end of program

The statements after // are the comments. They are not part of program. Following is the output
of the two cout statements in the main program.
Area of Rect1 = 2000
Area of Rect2 = 600
The above program is an example of object oriented program. It consists of class declaration
for objects which are rectangles. The name of class is Rectangle. Rect1 and Rect2 are two objects
(rectangles) of this class. See that function calls are linked with the objects on whose data they
operate, i.e., Rect1.Setsides(50, 40) and Rect1.Area(). The dot operator (.) provides the link. If
you simply call the function like Area() it will not work. So class is implemented only for its
objects. The various class functions are linked to the object for their execution.
 Introduction to C++ ❖ 15❖

INFORMATION HIDING
Classes also allow the hiding of the data members as well as some function members by declaring
them as private in the class declaration. This is illustrated in the above program in which int
length and int width are declared as private. Private members are accessible only through other
functions declared public in class declaration. Private data and functions are not directly accessible
even by the objects of the class. The objects of class may access its public data and function
members.
The class itself may be hidden from the user who only includes it (includes the name of class)
in the program and deals with the interface provided by public functions in the class like
Setsides () and Area () in the above example. In case of class program, a user of the program is
like a driver of a car who only knows about pedals (accelerator, clutch and brake) and steering
wheel and light switches which are the public interfaces or we may call them as public functions.
The driver does not know the complex mechanism behind the pedals and steering wheel. In fact,
a driver need not know all that just for driving the car. Similarly in a class program, the data and
some member functions may be totally hidden from the user of the program. Because after a
class code has been tested, debugged and verified we may store in a separate file. The class code
may not be available to user who only includes the relevant filename in his/her program. Besides
a class may be derived from another class which also may be having private members. The
intention is to hide as much as possible from the user of the program and classes provide the
mechanism for this. This is called information hiding.

1.11 OPERATOR OVERLOADING

In most of the programs, we often use operators such as +, –, /, etc. These are used for operations
like addition, subtraction, division, etc. Let us take operator +. This operator is defined to add
two integers or two decimal point numbers. Now if you want to add two vectors this operator
will not work because a vector has a magnitude and a direction. To overcome the problem of
direction, in vector mathematics, we add components of vectors. For example in case of two
dimensional vectors, x-component of vector A is added to x- component of vector B and y-
component of vector A is added to y-component of vector B. Here the vectors are shown by
bold letters. Let Ax and Ay be components of A and Bx and By be components of B. Let C be
the resultant vector with components Cx and Cy. So the addition of A and B is expressed as below.
Cx = Ax + Bx
Cy = Ay + By
We can, however, overload the operator + to carry out above additions, that is, we redefine
the functionality of + operator so that it will do the above two additions and will find Cx and
Cy.
With this facility we can write the following code in our program.
C=A+B;
This statement will do Ax + Bx and assign it to Cx and do Ay + By and assign it to Cy. Thus
we can do vector addition with the same operator +. This is called operator overloading.
❖ 16 ❖ Programming with C++

Similarly other operators may also be overloaded if so needed. You will yourself appreciate that
such a provision gives a tremendous extension to C++. There are many applications of operator
overloading such as in manipulation of vectors, complex numbers, matrices, any object that is
represented by more than one value can be dealt with the help of operator overloading.

1.12 INHERITANCE

We have already discussed that classes may be formed for the objects which have some common
characteristics. These characteristics are passed on to the classes that are derived from the base class.
The derived class inherits the base class without any modification while it may add more
specialized characteristics to the objects. The process is called inheritance. There is no end to the
process of inheritance. One may derive another class which inherits any one or more of the already
derived classes. Inheritance makes it possible to reuse an already tested and working class program.
Thus a big program may be divided into such classes which may be useful in the present program
as well as other programs. The Fig.1.9 illustrates the application of inheritance.
For sake of illustration let us consider two dimensional space. We all know that a line is made
of points, other shapes say polygons are made out of lines. The shapes like square, rectangle and
hexagon are polygonal shapes. Therefore, the functions of class Point are also useful in class Line.
Similarly the functions of class Line are useful in class Polygons and that of polygons are useful
in classes Square, Rectangle and Hexagon. Therefore, the upstream classes may simply be
inherited instead of repeating their code again and again in downstream classes.

Point class Point

Line class Line

class Polygons
Polygons

Square Rectangle Hexagon

class Square class Rectangle class Hexagon

Fig. 1.8: Inheritance

Thus inheritance helps in the reuse of the already established software. In the chain of
inheritances shown in Fig.1.8, the class Point is base class for class Line. Similarly the class Line
is base class for class Polygon which is also base class for class Square, class Rectangle, etc. A class
may have more than one class as base classes. An object of class Square is also an object of class
polygon but an object of class Polygon is not an object of class Square. This is because class
Polygon does not know about the classes downstream like class Square.

1.13 POLYMORPHISM

Polymorphism is a Greek word which means the ability to take different forms. In the previous
section we have seen how the operator + may be overloaded to carry out different operations.
 Introduction to C++ ❖ 17❖

This is an example of polymorphism. However, in such cases the choice of function is known
at the compile time itself. Therefore, this is called static binding or early binding. The run time
binding or late binding or dynamic binding is one in which the choice of appropriate function
is carried out during execution of the program (Fig.1.9). This is realised through declaration of
virtual functions in the base class and by use of base class pointer. A pointer is a variable whose
value is the address of another object to which it points.

Fig. 1.9: Polymorphism

1.14 NUMBER SYSTEMS

Since computer operations are ultimately connected to manipulation of numbers in the binary
form, it is worthwhile to know about different number systems which will be often used in the
programs. Below we deal with the number systems such as decimal, binary, octal and hexadecimal
systems. We normally use decimal system in our daily life calculations. This is a system with base
10 and digits are 0, 1, 2, 3, 4, 5, 6, 7, 8 and 9. Perhaps mankind started counted on fingers and
there are ten fingers so is the base of our number system. Suppose we have a number 456, it can
be written as below.
4 × 102 + 5 × 101 + 6 × 100 = 400 + 50 + 6 = 456
Here you see that apart from the value of digit, its position in the number also matters. In
456, the first digit 6 is at unit place and has value 6 while the digit 5 is at a place next to unit
and which has place value 10. Therefore, this 5 has value 5 ×101 = 50. Similarly the digit 4 is
at hundredth place has value 4 ×102 = 400
However, computer systems work on binary system. Binary system has a base 2. In a binary
number such as 1110111, the digit at unit place (right most) has place value 20 = 1. For the next
digit, the place value is 21 = 2. So if the digit is 0 its value is 0 but if the digit is 1 its value is 2.
Similarly the digit at hundredth place has place value = 1 ×22 , i.e. 4. In other words, digit 1 in
the hundredth place is equivalent to 4. Below we have tabulated the conversion of decimal
numbers from 0 to 10 to binary numbers.
Decimal number Binary number Conversion of binary into decimal

0 0 0 × 20 = 0
1 1 1 × 20 = 1 × 1 = 1
2 10 1 × 21 + 0 × 20 = 2 + 0 = 2
3 11 1 × 21 + 1 × 20 = 3
4 100 1 × 22 + 0 × 2 1 + 0 × 2 0 = 4
❖ 18 ❖ Programming with C++

5 101 1 × 22 + 0 × 21 + 1 × 20 = 5
6 110 1 × 22 + 1 × 21 + 0 × 20 = 4 + 2 = 6
7 111 1 × 22 + 1 × 21 + 1 × 20 = 4 + 2 + 1 = 7
8 1000 1 × 23 + 0 × 22 + 0 × 21 + 0 × 20 = 8
9 1001 1 × 23 + 0 × 22 + 0 × 21 + 1 × 20 = 9
10 1010 1 × 23 + 0 × 22 + 1 × 21 + 0 × 20 = 8 + 2 = 10

Numbers with base 8 and base 16 are also used. The numbers of digits in the different systems
are given in Table 1.1.
Table – 1.1

Number system Base Digits

Binary 2 0, 1
Octal 8 0, 1, 2, 3, 4, 5, 6, 7
Decimal 10 0, 1, 2, 3, 4, 5, 6, 7, 8, 9
Hexadecimal 16 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
The digits A, B, C, D, E and F in hexadecimal system are equivalent to 10, 11, 12, 13, 14
and 15 respectively in the decimal system. A number such as 0000000111001000 may be
converted back to decimal number as given below.
1 × 28 + 1 × 27 + 1×26 + 0 × 25 + 0 × 24 + 1 × 23 + 0 × 22 + 0 × 21 +
0 × 20 = 256 + 128 + 64 + 8 = 456
In the above exercise it is shown that each place in the binary number has a power of the
base 2. Thus for the digit on the extreme right the power is zero, so for value it is multiplied
by 20 or 1. For the next digit the power is 1, so its value is obtained by multiplying it by 21,
and so on.
Figure 1.10 illustrates the conversion of a decimal number into a binary number.

2 456
2 228 – 0
2 114 – 0
2 57 – 0
2 28 – 1
2 14 – 0
2 7–0
2 3–1
2 1–1
0–1

The number 456 when written in binary would look as this 1 1 1 0 0 1 0 0 0

Fig. 1.10: Converting a decimal number into binary

 Introduction to C++ ❖ 19❖

Fig. 1.11
The same number 456 may also be written in octal or hexadecimal system as well. We repeat
the exercise of Fig.1.10 for octal and hexadecimal systems as shown in Fig.(1.11a) and Fig. (1.11
b) respectively. The following table shows the number 456 in different number systems.
Table – 1.2

Number system Base Representations of decimal number 456

Binary 2 111001000
Octal 8 710
Decimal 10 456
Hexadecimal 16 1C8

In a computer program the octal numbers are preceded by 0 (zero) and hexadecimal numbers
are preceded by 0x or 0X. Thus 456, in octal is 0710 and in hexadecimal it is written as 0X1C8.
For conversion from binary to octal or hexadecimal see the following grouping of digits.
Binary equivalent of 456 111 001 000
Octal 7 1 0
Binary equivalent of 456 1 1100 1000
Hexadecimal 1 C 8
For converting binary into octal, make groups of binary digits with three digits in each group
starting from extreme right. The decimal digit represented by each group is the digit in octal
representation. In the above example, starting from right, the first group of three binary digits
are all zero. So the octal digit is also zero. The second group 001 is equivalent to octal 1. The
third group 111 is equivalent to 7. So the number in octal representation is 0710.
Similarly for the hexadecimal representation from binary, make groups of four binary digits
as illustrated above. Make a hexadecimal digit out of the four binary digits. The right most group
(1000) evaluates to 8. The next group 1100 evaluates to 12 in decimal which is C in hexadecimal.
The last group is only 1. So the number 456 in hexadecimal representation is 0X1C8 or 0x1c8.
In a program, decimal numbers should not be preceded by zero ( 0).

FRACTIONAL NUMBERS
In a decimal system a fractional or a decimal point number say 456.25 may be written as
456.25 = 4 × 102 + 5 × 101 + 6 × 100 + 2 × 10–1 + 5 × 10–2
Same method applies to binary fractional numbers. Say we have a number
❖ 20 ❖ Programming with C++

x = 111001000.01
in binary. It may be converted to decimal system as below.
x = 1 × 28 + 1 × 27 + 1 × 26 + 0 × 25 + 0 × 24 + 1 × 23 + 0 × 22
+ 0 × 21 + 0 × 20 + 0 × 2–1 + 1 × 2–2
= 256 + 128 + 64 + 8 + 0 + 0.25 = 456.25
For converting 0.25 from decimal system to binary, we have to successively multiply it by
2 till it is 1 or non-fractional as below.
0.25 × 2 × 2 = 1
Therefore, the number 0.25 in decimal system is equal to 0.01 in binary.
Exercise – Convert the fraction 0.1, 0.2 and 0.5 from decimal system to binary.
Solution
Conversion of 0.1—Multiply 0.1 successively by 2, till it is 1 or just more than 1.
Binary fractional number
.1 × 2 = .2 less than 1 .0
.1 × 2× 2 = .4 less than 1 .00
.1 × 2 × 2 × 2 = .8 still less than 1 .000
.1 × 2 × 2 × 2 × 2 = 1.6 now multiply fraction .6 as below .0001
.6 × 2 = 1.2 now multiply fraction .2 as below .00011
.2 × 2 × 2 × 2 = 1.2 now multiply fraction .2 by .00011001
Therefore .1 of decimal = .00011001 of binary. As shown above the fraction could still continue.
Conversion of 0.2 of decimal into binary
0.2 × 2 × 2 × 2 = 1.6 fraction part now left is .6 .001
.6 × 2 = 1.2 fraction part now left is .2 .0011
.2 × 2 × 2 × 2 = 1.6 fraction part now left is .6 .0011001
Therefore .2 of decimal = .0011001 of binary.
Similarly it can be shown that .5 of decimal = .1 of binary.

1.15 BITS AND BYTES

The smallest unit of computer memory is a bit, which may have values either 0 or 1. In
computer we often deal with groups of bits. A group of 8 bits is called a byte. The characteristics
of a memory is that if a bit has been set to one it will continue to remain in that state until it is
reset to zero. And if it is reset to 0 it would remain so unless it is set to 1. A simple example is
that of a switch. When you press the switch, it switches on a light, and it continues to remain
on untill you put it off or reset it. When a switch is put on it connects a high voltage to the light
circuit. We may call it as state 1. When it is put off the circuit voltage becomes zero. We may
call it a state 0. In electronics there are circuits which are called flip-flops. A flip-flop is just like
a switch. When it has high voltage (≈ 5 V) on its output point we say it is set and when the
voltage at its output terminal is 0 we say it is at zero or reset.
 Introduction to C++ ❖ 21❖

The output state can be manipulated by input signal to the flip-flop. Thus we can make its
output 0 or 1 as we wish just like a switch which can be put on or put off, the difference being
that in case of flip-flop it is done by electronic circuit while in case of manual switch it is done
manually. We have qualified the switch as manual because switching can also be done
electronically. Each flip-flop is one bit of memory. A bit can have value 0 or 1. Now you
understand why binary numbers are used in computers. Binary numbers also have only two
digits, one is 0 and second is 1. Computer memory also has two states 0 or low and 1 or high.
If we have an array of such flip-flops, some set high or at 1 and some set to low or at 0 at their
outputs, we can represent a binary number. For instance, Fig.1.12 shows four bits. They have
the place values from unit for the bit on extreme right to 8 for bit on the extreme left. If all the
bits are set (have value 1) the total number stored is equal to 8 + 4 + 2 + 1 = 15 in decimal
system.

Bit number 3 2 1 0
Binary number (when set) 1 1 1 1
Place value 23 =8 22 =4 21 =2 20 =1
Number stored when all the bits are set = 1 × 8 + 1 × 4 + 1 × 2 + 1 × 1 = 15
Fig. 1.12

Below in Fig. 1.13 we show 4 flip-flops by four squares numbering from 0 to 3. In computer
terminology the counting starts from 0. Here we take right most bit (Fig.1.13) as the 0th bit at
the unit place. In programming also as you will see in the chapter on arrays that the first member
of an array is the 0th element. So it is better that you make a habit to count from 0. The next
flip-flop (shown by square at 1) is at place where place value is 2. The flip-flop next to it (3rd
from right and numbered as 2) has a place value 22 = 4 and the last has place value 23 = 8.
Now if the bit 0, 1 and 3 are set (each equal to 1) and bit 2 is 0. The number stored is 11.
It is explained below. 3 2 1 0

Output state 1 0 1 1
Number held 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 = 8 + 0 + 2 + 1 = 11

Fig. 1.13

Starting from right of Fig.1.13, the 0th bit is high representing 1, the next one is also high but
it is one digit to the left of 0th bit. So here 0 is zero, but 1 is equal to 1 × 21 = 2. Similarly at the
third place it is zero so it is equal to 0 × 22 = 0. At the fourth place the bit is high so it represents
1 × 23 = 8. So the total of all these is 11. Similarly, if all of them were high they would represent
a number 15, and when all of them are low they represent 0. So a set of 4 memory bits can hold
any number from 0 to 15 or from 0 to (24 – 1). Similarly, if we have 8 bits of memory arranged
as shown in Fig.1.14, we can store any number from 0 to ( 28 – 1 ) = 0 to 255.
❖ 22 ❖ Programming with C++

One byte

Bit No. 7 6 5 4 3 2 1 0

1 1 1 1 1 1 1 1

Place value 27 = 128 26 = 64 25 = 32 24 = 16 23 = 8 22 = 4 21 = 2 20 = 1

Total = 128 + 64 + 32 + 16 + 8 + 4 +2 + 1 = 255

Fig. 1.14: One byte can store any positive number from 0 to 255

So one byte can hold any positive number (unsigned number) from 0 to 255. Similarly if
the number is unsigned (positive) 2, 3 and 4 bytes can hold numbers as given below.
2 bytes (16 bits) can hold an unsigned number from 0 to (216 – 1) = 0 to 65535
3 bytes (24 bits) can hold any unsigned number from 0 to (224 – 1) = 0 to 16,777,215
4 bytes (32 bits) can hold any unsigned number from 0 to (232 – 1) = 0 to 4,294,967,295
Often we have to deal with positive as well as negative numbers (signed numbers). In such
cases, the left most bit of the memory block allocated to the number is reserved for sign. If this
bit is 0 the number is positive and if the bit is set to 1 the number is negative. So for storage of
signed number we now have one bit less. Therefore, for signed numbers the maximum value
of number that can be stored on a n-bit memory becomes (2n-1 – 1). Thus the ranges of signed
numbers that can be stored on 1, 2, 3 and 4 bytes are as follows.
Range of signed number that 1 byte can hold
= – 28–1 to (28–1 – 1) , i.e. – 128 to 127
Range of signed number that 2 byte can hold
= – 216–1 to (216–1 – 1) = – 32768 to 32767
Range of signed number that 3 byte can hold
= – 224–1 to (224–1 – 1) = – 8388608 to 8388607
Range of signed number that 4 byte can hold
= – 232–1 to (232–1 – 1) = – 2147483648 to 2147483647
In general, characters such a ‘A’, ‘b’, etc., are stored in one byte, short integers (whole
numbers) are stored in two bytes, integers are stored in 4 bytes. The floating decimal numbers
are stored in 4 bytes for single precision (float) and in 8 bytes for decimal point numbers with
double precision.
 Introduction to C++ ❖ 23❖

1.16 COMPUTER PERFORMANCE

Computer performance is generally measured by how quickly it performs a task. If it takes more
time then its performance is low. The speed of CPU (frequency of its oscillator) is one factor
that contributes to the efficiency. But processor does not work alone. It works with support of
other electronic chips whose speed is equally important. Besides, the type of program that it is
handling, the language of program, etc., all affect its performance. Among the devices that support
CPU, the most important ones are the memory devices in which the processor stores the data
and extracts data from. Generally the speed of memory devices is much lower than the speed
of the processor and this low speed becomes a drag on the processor because it has to wait till
the process of storing data or process of extracting data is completed. Fig.1.15 shows the different
memory devices which are used by the processor to carry out a program. Different memory
devices have different speeds and hence the cost (Fig.1.16). The performance of a memory device
is measured by the access time. This is the time taken by memory for extraction of a data. The
fastest memory which is closest to the processor is L1 cache memory which in some processors
is on the processor chip itself. It has the minimum access time of the order of 1 to 2 nano-seconds
(1 nano-second =10–9 sec.). Then comes the L2 cache memory that may be on a separate chip.
Its access time is nearly twice that of L1. Then comes the RAM the main memory which is
extensively used by processor. The cache memories are also part of RAM. Speeds and prices of
memory devices vary over a good range. The slowest are the bulk storage memories such as hard
disc, magnetic tape etc., and their access time varies from seconds to minutes. Fig.1.16 gives an
idea of the access times of different memory devices.

Floppy Disc

Fig. 1.15: Types of memories in a computer

❖ 24 ❖ Programming with C++

L1 cache
1 – 10 ns

L2 cache
20 – 30 ns
cost

Hard disc
25 – 120 ns
CD ROM
Tape
100 000 ns

in mins

20 40 60 80 100 120 100000ns 1 min

Access time

Fig. 1.16

It is logical to infer that a computer with large L1 and L2 would perform better. Besides L1
and L2, the computer performance also depends on RAM and the quality of program. A
program is more efficient if it generates less machine code. In this respect the programs written
in machine language are the fastest because there is nothing between the program and the
processor. The programs written in assembly language come next , i.e. take more time as these
have to go through the assembler. The programs written in a high level language have to be
converted into machine language by a compiler or interpreter. Naturally these programs take
more time than those written in assembly language. Programs in C++, when compared with
other high level languages on efficiency scale, come midway, nearly as good as java.

EXERCISES
1. What do you understand by following terms?
(i) Machine language
(ii) Assembly language
(iii) High level languages
2. What are the major additions from C to C++?
3. What is a compiler? Why is it needed to run a program written in C++?
4. What is a class?
5. What are the different types of computer memories?
6. Explain the following number systems.
(i) Decimal (ii) Binary
(iii) Octal (iv) Hexadecimal
7. Convert decimal number 46535 into binary, octal and hexadecimal numbers.
 Introduction to C++ ❖ 25❖
8. Convert the following hexadecimal numbers into binary numbers.
(i) d368c
(ii) f 0 abc7
(iii) 368c
9. Convert the following binary numbers into hexadecimal numbers.
(i) 1110101011001
(ii) 1010110001101.10101

10. Convert the following decimal numbers into binary numbers.

(i) 65784
(ii) 100765
(iii) 133.25
11. Convert the following binary numbers into decimal numbers.
(i) 11001100110
(ii) 10011110001.111
(iii) 11110111111
12. How are the hexadecimal numbers differentiated from octal numbers in output from a computer?
13. What do you understand by the following terms?
(i) Microprocessor (ii) Compiler
(iii) Assembler (iv) Interpreter
14. What do you understand by the following?
(i) RAM (ii) ROM
(iii) L1 cache memory (iv) L2 cache memory
15. What do you understand by the following terms?
(i) Flip flop
(ii) Bit
(iii) Byte
16. Explain the following terms.
(i) Inheritance
(ii) Operator overloading
(iii) Polymorphism
17. What do you understand by object oriented programming?
18. How are binary numbers stored in a computer?
19. What are classes and objects?
20. How big a number may be stored in 3 bytes if it is one of the following?
(i) signed number (ii) unsigned number
21. A class program is to be made for determining area of any polygon. What data is required for
the purpose?
❖ 26 ❖ Programming with C++
22. Distinguish between the following styles of programming.
(i) Procedural programming.
(ii) Modular programming.
(iii) Object oriented programming.
23. Distinguish between the following terms.
(i) Class and object
(ii) Base class and derived class
(iii) Data abstraction.
24. What data should be printed on a driving licence of a person? (Hint: use data abstraction)
25. What data should be printed on the identity cards of students of a technical institute?
26. The city police decides that all domestic servants should carry an identity card. What data
do you think should be printed on it?
27. The employees of a bank are required to wear an identity card during working hours in the bank.
What data do you think should be printed on it?

❍❍❍
CHAPTER

++

2.1 INTRODUCTION

A typical program in C++ may comprise a list of statements involving variables (objects whose
values may change during the execution of program), constants (whose values do not change),
operators like +, –, etc. and functions, etc. Computer recognises them by their names just like
you are recognised by your name. Obviously, no two variables or constants should have same
name. However, functions may have same name if their parameters are different. It must be
emphasised here that C++ is a case sensitive language, which, means that it will take ‘A’ and
‘a’ as two different objects. Similarly Area, area and AREA are three different objects. Therefore,
while writing a program the name of a variable, constant or a function should be written
consistently in the same fashion throughout the program.
Like any other high level programming language, C++ also has certain rules, special syntax
and keywords to help the programmers to write a program and execute it. Keywords have special
meanings for the compiler and are used to control and execute the program. Naturally, these
words should be used in a program only for the purpose they are meant for. Their use as names
of variables, constants, functions or objects will create errors in the program. The keywords used
in C++ are listed in Table 3.2 in Chapter 3. Besides these, there are a number of files called
header files and functions in C++ Standard Library which help in the execution of the
programs. Therefore, every program has to include the header files which are required by the
program.
The keywords as well as names of files and functions in C++ Standard Library are in general
defined in lower case. As already mentioned the keywords should not be used as names, however,
if the case of any letter in a keyword is changed it is no longer a keyword. Thus a safe bet is that
the starting letter in a name may be made capital.
Also C++ is highly typed language which means that the data is categorized into different
types. For example, whole numbers form a category called integers. So when a variable whose
value can only be in whole numbers, is declared we write int (short form of integer) before its
name. The int is its type. Variables which have values in floating decimal point numbers such
as 2.4 or 3.14159, etc. form another category. For declaration of such variables we write float
or double before their names. For variables which have values in form of characters are of type
char.
❖ 28 ❖ Programming with C++

An interactive program requires the user of the program to put in some data during the
execution of the program. The data may comprise numerical values, or a string of characters
which may be a statement or simply yes or no. In the following we first learn the essential tokens
of a C++ program which are a must in every program.

2.2 COMPONENTS OF A SIMPLE C++ PROGRAM

Program 2.1 given below, illustrates the essential components or tokens of a C++ program. In
this program it is desired to put two sentences or messages on the output device. If the output
device is not specified in the program, the monitor of the computer is the default output device.
In such a case, the output will be displayed on the monitor. Monitor is connected to one of the
output ports of the computer. Similarly keyboard is the default input device.

PROGRAM 2.1 – A sample program which displays two messages on monitor.

#include <iostream.h>
int main()
{ std::cout<<“Hello, Welcome to programming with C++!\n” ;
std::cout <<“Are you interested to join?”<< std::endl;
return 0 ;
}

HEADER FILES
The first line of the Program 2.1 is #include <iostream.h> in which the symbol # is a pre-
processor directive. It is a signal for pre-processor which runs before the compiler. The statement
directs the compiler to include the header file <iostream.h> (input/output stream) from C++
Standard Library in this program. The file <iostream.h> has the procedural software in the form
of functions which are used with its two objects cin and cout for performing input and output
respectively. The cin is used along with extraction operator (>>) for input and cout is used along
with insertion operator (<<) for output. The name of a header file when included in a program
is enclosed between a pair of angular brackets < >. The header file name may as well be enclosed
between double quotes “ ”. When the header file name is enclosed between angular brackets the
pre-processor searches for it only in C++ Standard Library. The angular brackets indicate to pre-
processor that it is part of C++ Standard Library. When it is enclosed between double quotes
(“ ”), the pre-processor first searches the directory in which the file containing the directive
#include appears. If this search is not successful, the search goes to C++ Standard Library.
Therefore, generally for a user defined header files the name is written in double quotes and for
the C++ Standard Library header file, the name is written in angular brackets.
Any program requiring standard input, i.e. from keyboard, and standard output, i.e. to
monitor, must include the header file <iostream.h>. You may include other header files as well
from C++ Standard Library if your program requires them. For example, if you want to carry
out evaluation of mathematical functions like square-root, sin(θ), cos (θ) or log (6), etc., you
must include the header file <cmath>. The header file cmath has the software for evaluation of
many such functions. For including more than one header file in a program, these should be
written in successive lines as illustrated below. Only one header file is written in one line.
 Structure of a C ++ Program ❖ 29❖
#include <iostream>
#include <cmath>
#include <cstdlib>
Also note that there is no comma or semicolon or full stop anywhere in any of the above
three lines or at the end of lines. The earlier convention of writing the header file with extension
(.h) such as <iostream.h> is now not recommended and is written simply <iostream>. Similarly
the other header files such as <math.h> and <stdlib.h> are now written as <cmath> and <cstdlib>.
Both versions are supported by compilers. The difference between the two is that header file
having names like <xxxx. h> have their names in global namespace while those without (.h) have
their names in namespace std which is being increasingly used now. The term namespace std and
its use is explained latter in this chapter.

INT MAIN( ) OR VOID MAIN( )

The second line of the above program (int main()) indicates the beginning of the program. In
C++ the parentheses ( ) are used to indicate that the identifier or name on its left is a function.
Here main () is also a function. Every program in C++ must have only one main() function.
The word int stands for integer. Here int main() means that function main() is a return type
function which returns an integer value. In case of a return type function the last statement is
return value; . In the present case the last line of the program reads return 0;. If the above
program runs successfully, it will return 0 to the system. For some compilers the statement return
0; is optional but if you are working with Microsoft C++ Visual 6 compiler, this statement is
required, otherwise, the compiler will show error. The code lines after int main() and enclosed
between the left brace ‘{’ and right brace ‘}’ form the body of the program which may have
several lines of code. The left brace ‘{’ is placed at the beginning and the right brace ‘}’ is placed
at the end of a program. In general, a program may have several pairs of these braces in its body
as well. These are used for various purposes, however, the above two braces are essential.
A typical program in C++ is illustrated below in Fig. 2.1. It starts with names of header files
to be included in the program. These are followed by function int main () or void main () and
then the body of program enclosed in the curly braces{}. These braces generally contain
expressions, statements and operators with a purpose to carry out some action on the data
presented to the program. Expressions are the conditions if fulfilled the statement following the
expression is carried out. In the program sample you will observe that all the statements in a
C++ program end with a semicolon (;). Besides the statements, the program may contain
conditional expressions like the one illustrated in Fig.2.1, i.e., if(expression). Such expressions
do not end with semicolon (;).
# include <iostream>
int main ()
{Statement;
if(expression)
{Statement;
————;
}
Statement; // comment
return 0;
}
Fig. 2.1: A typical program statements
❖ 30 ❖ Programming with C++

The body of Program 2.1 comprises two output statements and the line return 0;. The last
line has already been discussed above. The output statements are discussed below.

OUTPUT STATEMENT – USE OF COUT

The first output statement of Program 2.1 is repeated below for ready reference.
std::cout<<“ Hello, Welcome to programming with C++!\n” ;
The C++ standard library functions and classes are defined under namespace std. The operator
(::) is the scope resolution operator. The line std :: cout means that we are calling an object cout
which is defined under namespace std or which belongs to header file <iostream.h> of C++
Standard Library. The object cout is used with the stream insertion operator (<<). During the
execution of the program the operator << inserts the argument on its right hand side into the
output stream. In the present program the statement on the right of operator << is a string of
characters enclosed between double quotes (“ ”). A string of characters as defined by ASCII (see
Appendix A) and enclosed between double quotation marks is put to output stream as it is
written. Even if it contains a keyword or digit or any other symbol, it would go to output stream
as it is written without any change.

2.3 ESCAPE SEQUENCES

Some characters preceded by back slash character (\) like the one at the end of the above output
statement, i.e. “\n”, have a special meaning for the compiler. Back slash (\) is an escape character.
The combination “\n” represents an escape sequence which directs the cursor to new line. Thus
“\n” is new line character. It directs the cursor to next line, naturally any statement on the right
hand side of “\n” will go to next line. But, \n has to be enclosed between double quotes like “\n”.
In some situations it may not be possible to put this symbol along with other statement as it
has been done above, and, if it is desired that the output should now shift to next line, then it
may be written in a separate line as illustrated below.
std::cout<<“\n”;

Similarly, a back slash followed by character ‘t’, i.e. , “\t” represents an escape sequence which
moves the cursor to right by certain spaces in the same line. The number of spaces for the shift
may be modified by adjusting the setting. The other such escape sequences and their actions are
described in Table 2.1 below.
Table 2.1 – Escape sequences

Escape character Description of action

\a Bell or beep is generated by the computer on program execution.
\b Back space. Moves cursor to previous position.
\f Form feed. Advances cursor to next page.
\n Shift to new line.
\r Carriage return. Positions the cursor to the beginning of current line.
Contd...
 Structure of a C ++ Program ❖ 31❖

\t Horizontal tab. Moves the cursor by a number of spaces or to next tab stop
in the same line.
\v Vertical tab.
\\ Displays a back slash character (\).
\’ Displays a single quote character ( ’).
\” Displays a double quote character (”).
\? Displays a question mark (?).
\0 Null termination character. Signifies end of a character string.
\o Code for octal numbers.
\x Code for hexadecimal numbers respectively.

The next line in Program 2.1 is also an output statement, enclosed in double quotes. After
the quotes is the end of line code (std::endl;). In this code we have used scope resolution operator
(::) because we are calling endl which belongs to C++ Standard Library. The effect of endl is similar
to that of “\n”, i.e. it directs the cursor to new line. Thus what follows endl goes to the next line.
Both the “\n” and std::endl may be used any number of times in the same code line. The output
will correspondingly go to different lines.
The last line in the Program 2.1 is return 0; and it has already been explained above. The
expected output of the Program 2.1 is the display of two messages on the monitor as given
below.
Hello, Welcome to programming with C++!
Are you interested to join?
When a function does not return any numeric value as in the present case of main() we may
as well write its type as void. Thus we could have written void main() in place of int
main() in the above program. But in that case the statement (return 0;) is not required.
This applies to any other void function as well. However, if we do not specify int or void
before main() compiler would take it as int by default.

COMMENTS
The comments are not part of a program but are included so that the programmer can later
remember what the program and its various variables stand for. It is a good practice to include
comments, particularly for big programs which involve a large number of variables. A week after
writing the program, the programmer may forget what the program does or what the different
variables stand for. The comments help not only in recollecting the program written a few days
ago, they also help even during writing a program particularly if it is a big program.
A comment may be introduced by first putting double slash (//) followed by the comment
up to end of the line. It can be put anywhere in the program. The compiler simply neglects what
follows the double slash (//) up to the end of line. Therefore, a comment may be put in the same
line as the code or in a new line but after the double slash (//). If the comment is long and goes
to next line, another double slash (//) is needed before the next line. Alternatively, a long comment
❖ 32 ❖ Programming with C++

may as well be enclosed between the C type comment symbols, i.e. /* before the start of
comment and */ at the end of the comment. Both these methods are illustrated in Program 2.2.

PROGRAM 2.2 – It illustrates how to include comments and use of void main () in place
of int main().

// A program with comments

#include <iostream>
void main()

//Anything written after double slash is neglected up to the

//end of line. The comment can be put anywhere in the program.

{ std::cout << “Hello, Welcome to programming in C++! \n”;

/* Notice the blank spaces in the beginning of the first two lines, and also
the missing lines. All these blank spaces are neglected by compiler. Comments
may also be put in the c-style, i.e. between the symbols/ * and */ as done in
this comment.*/
std::cout <<“Are you interested?”<<std::endl;
/* When you use void main() do not include the statement ‘return 0;’, because
the void functions do not return any value*/
//and compiler will show it as an error if return statement is included.
}

In the above program there are too many comments and program code is submerged in
them, so the code lines are made bold so that these are distinctly visible. The first line is a
comment. The output is same as for Program 2.1. Notice that blank spaces are introduced in
the first three lines. Also after void main () in program body a line is blank. All these are called
white spaces and are neglected by the compiler. Therefore, you may write the program as you
like so that it is easily readable. There is no requirement that every line must start from the
beginning. In above program int main()is replaced by void main(). Since void functions are non-
return type functions so the statement return 0; is not included in the program.
The Program 2.3 given below illustrates application of some of the escape sequences listed
in Table 2.1. Remember that they have to be enclosed in double quotes (“ ”).

PROGRAM 2.3 – Illustrates application of some escape sequences.

#include <iostream>
int main()
{
std::cout<<“Hello,\a\’Welcome to programming with C++!\’\n”;
 Structure of a C ++ Program ❖ 33❖
/* Inclusion of character “\a” will cause the computer to generate a sound
(a beep) when program is executed */

std::cout<<“Hello,\\ Welcome to programming with C++!\\\n”;

// This prints \ see 2nd line of output.

std::cout<<“ Welcome to programming with C++!\rHello\n”;

// inclusion of \rHello brings Hello to the front of line.

Std::cout<<“Hello\t Welcome to programming with \t\”C++!\”\n”;

// The character \” puts double quotes to the statement

return 0;
}

The expected output is given below.

Hello,’Welcome to programming with C++!’
Hello,\ Welcome to programming with C++!\
Hello Welcome to programming with C++!
Hello Welcome to programming with “C++!”
The program has been made self explanatory with the help of comments included in the
program. You may verify them from the output. In implementing the first line of code after int
main(), the computer makes a sound like a beep. In the second line of output slashes are put.
In the third line of output the word Hello has been bought forward to the beginning of line.
But you have to keep space for it otherwise it will replace some characters of welcome. In the
4th line of output the application of tab “\t” has shifted the cursor by a few spaces. So in the
output we find space after Hello and after with. Also note that C++! has been enclosed in double
quotes.
THE CIN

In many interactive programs the user is required to put in some data. In the following program
we make use of function cin (the standard input stream) for this purpose. This is used along with
the operator >> which is called extraction operator. This is followed by the name of variable
in whose memory block the data has to be stored. The input data should be of same type as that
of the variable declared. The operator >> directs the data to cin. After typing the data for cin,
do press the ‘enter-key’, otherwise, the computer will not proceed further. The following
program illustrates the application of cin.

PROGRAM 2.4 – Application of cin in a user interactive program.

#include<iostream>
int main()
{
❖ 34 ❖ Programming with C++
int length =0; // length is the name of variable of type int
int width=0, area=0; //width and area are also names of variables.
std::cout<<“Write the length and width of a rectangle: ” ;
// output statement
std::cin>> length>>width; // Input statement
area = length*width; // computation of area

std::cout <<“length = ” <<length<<“\t width = “<<width<<“\t area = “

<<area<<endl; // output statement

return 0 ;}

You should note that in the above program all the code lines except the first two end with
semicolon. The first line comprises the header file and the second is int main(). A common mistake
by a student involves use of semicolon at wrong places. You would also observe that all the code
lines in the above program do not start from extreme left. As already mentioned above, in C++
the white spaces are neglected. So it does not matter where you start the line. The program may
be written in a way so that it is easier to read. The operation of above program is explained below.

2.4 VARIABLE DECLARATION AND MEMORY ALLOCATION

In Program 2.4, the second code line after int main() declares an integer variable by name ‘length’.
The word length is identifier or name of the variable, just as you are identified by your name.
When you declare a variable, a block of memory is allocated for it for putting its value. In this
case 0 is assigned to it. The size of the memory block depends on the type of variable (see Chapter
3 for details). The declaration of a variable in a program is done by first writing its type, give
space then write the name or identifier for the variable as illustrated below.
type identifier ;
This is illustrated below.
int n ; // The statement declares an integer with name n.
The details of fundamental types in C++ are discussed in detail in Chapter 3. Here we may
mention that the five types of data described in Table 2.2 below are the most common in C++.
Examples of their use are also given in the table.
Table 2.2* – Most commonly used data types

Type Description Examples of declaration

void no data type void main()
(used with functions & pointers) void F();
int whole number or integer int length;
int k = 6;
float decimal point number. float y;
(single precision) float PI =3.141;
Contd...
 Structure of a C ++ Program ❖ 35❖

double decimal point number double A;

(double precision) double PI = 3.141592653;
char all alphabets like A, b, C char ch;
along with symbols as char x = ‘A’;
defined in ASCII code Here ch and x are names of variables.
(see Appendix A) ‘A’ is the value assigned to variable x.
* See Chapter 3 for more details.

In the 5th line of Program 2.4 two more variables are declared with names ‘width’ and ‘area’.
You may declare any number of variables of same type in a single line. The type should be
mentioned at the start of line and the variable names should be separated by comma (,).
Before a variable is used in the program it should be initialized which means that a value
should be assigned to it, otherwise the complier will show error or warning. To start with we
have assigned 0. As already mentioned above when a variable is declared a block of memory is
allocated for storing its value. The value may be either assigned by the programmer or calculated
during the execution of program. With the declaration of three variables, i.e. length, width and
area the compiler allocates three blocks of memory. A graphical illustration of this is given in
the Fig.2.2.
length width area

0 0 0

Fig. 2.2: Allocation of memory blocks on declaration of variables

At the time of declaration of variables in Program 2.4 we assigned 0 values to length, width
and area, so the allocated memory blocks show 0. The actual values are entered by the user when
the program runs (user interactive program). On clicking to run the program it displays following
sentence on the monitor.
Write the length and width of a rectangle:
At the end of line you will find a blinking cursor. Here you have to enter two numbers. You
may type one say 15, then give a space and type second number. In this case, 4 has been typed.
After typing, the line would look like as given below.
Write the length and width of a rectangle: 15 4.
After typing 15 and 4 press the ‘enter-key’, with this the extraction operator (>>) will direct
the two values to std::cin which will place them in the memory blocks allocated for them when
they were first defined. The number 15 will be placed in the memory block allocated for length
and number 4 will be placed in the memory block allocated for width. The Fig. 2.2 now changes
to the following.
❖ 36 ❖ Programming with C++

length width area

15 4 0

Fig. 2.3: Condition of allocated memory blocks after completion of cin function

In the next code line of the program the two numbers are multiplied and the product is
assigned to area. Assignment is carried out with operator ( = ). On the left side of the operator
we write name of variable and on its right side the value to be assigned. This line of program is
processed by three operations, i.e. (i) the numbers 15 and 4 are copied from their locations then
(ii) these are multiplied and (iii) the product is placed at the memory location reserved for area.
After the variable area is assigned the product (60), the 0 is replaced by 60. The Fig. 2.3 now
changes to the following figure.
length width area

15 4 60

Fig. 2.4: Condition of allocated memory blocks after assignment of product to area

The next code line in the Program 2.4 is the output statement. For execution of this line
the numbers from the three locations are copied and displayed on the monitor. The words in
double quotes are displayed as they are typed in the program. Values of variables are put where
their names appear in the cout expression of the program. So you get the display as given below.
Write the length and width of a rectangle: 15 4
length = 15 width = 4 area = 60

2.5 NAMESPACES

Namespaces is a powerful feature of C++ language. It enables the programmer to group a set of
objects or functions under one name. All the classes, objects and functions of C++ Standard
Library are defined under namespace std. We can simplify the writing of std :: cout to cout and
std :: cin to cin by including the following directive in the program.
using namespace std;
Note that this line ends with a semicolon. The namespaces are useful when you are
combining two or more parts of a program and you are afraid that some of names of different
variables may be same in the two parts. To distinguish that a particular set of names belongs to
a particular part of program we make use of namespace. Say we may put the declaration
namespace XA
in the beginning of part A and similarly put
namespace XB
 Structure of a C ++ Program ❖ 37❖
for the part B. Now if the two are combined and we want to access a variable n of program part
A we can write XA :: n. Similarly if part B of the program also has a variable by name n we can
access it through scope resolution operator as XB::n . Even though the name is same ( n ) the
values in two cases would be as assigned in XA and XB respectively.
More details of using namespace are discussed in Chapter 20. Here we contend with that
by adding using namespace std; in the beginning of the program, the std:: is no longer needed
with cout and cin and other names belonging to C++ Standard Library such as endl, etc. The
following program illustrates the application of namespaces and the directive ‘using namespace
std’. First, a group of names n, m, k and R are declared under namespace NS1. The same names
are again declared under namespace NS2. The two sets of name are used in a program with
application of scope resolution operator :: or by adding the statements such as,
using namespace NS2;

Under this statement the names belonging to NS2 can be used without the scope resolution
operator but the names belonging to NS1 will have to use NS1:: before their names.

PROGRAM 2.5 – Illustrates application of namespaces.

#include <iostream>
using namespace std; // use of std namespace
namespace NS1 //no semicolon at end of line
{ int n = 3;
float m = 2.5;
int k = 2;
double R = n*m*k; }
namespace NS2 //no semicolon at end of line
{float n = 4.0; //Names are same as in NS1 but values are
// different.
int m = 2 ;
double k = 3.0 ;
double R = n*m*k; }
int main()
{ int Square;
int Product ;
using namespace NS2;
Square = n*n + m*m ; // values under NS2 are used
Product = NS1::k * NS2::m; // k belongs to NS1 and m to NS2
cout << “Square = “ << Square << “,\t Product = “ << Product<<endl;
cout<< “ R = ”<< NS1::R <<endl; // This R belongs to NS1
cout << “ R = ” <<NS2::R<< endl; // This R belongs to NS2
return 0 ;
}
❖ 38 ❖ Programming with C++

The output is as under. The output is made self explanatory by using comments.
Square = 20, Product = 4
R = 15
R = 24

2.6 USER INTERACTIVE PROGRAMS

In a user interactive program the user is required to feed the data asked by the program. Program
2.4 discussed above is an example of user interactive program. The Program 2.6 given below is
yet another example of the same. The program makes use of the object cin of <iostream> header
file. The difference between the two programs, i.e. 2.4 and 2.6 is that in Program 2.6 we have
used the directive using namespace std; which simplifies the writing of names belonging to C++
Standard Library, for example, instead of writing std :: cout we simply write the code as cout.

PROGRAM 2.6 – This program illustrates the application of cin and use of namespace std.

#include <iostream>
using namespace std;
int main()
{ int D;
float PI ;
// Here D stands for diameter and PI stands for π.
double A, C;
// A stands for area and C stands for circumference.
cout<<“Write values of D and PI:”;
cin>> D>>PI; //statement for input of D and PI
cout<<“You have written the values as D = “<<D<<“ and PI = “ <<PI<<endl;
// endl may be used in place of “\n”
A = PI*D*D/4; // computation of area
C = PI*D; // computation of circumference
cout <<“D = “<< D <<“, A= “<< A <<“ and C= “ <<C <<endl;
return 0;
}

On clicking to run the program, the following statement is displayed on the monitor.
Write values of D and PI:
You will see a blinking cursor at the end of line. There you type the value of diameter, in
this case it is typed 10. Give a space and write value of PI and press enter-key. The program
will not proceed further if you do not press enter-key. The output of the program as displayed
on monitor is given below.
 Structure of a C ++ Program ❖ 39❖
Write values of D and PI:10 3.14159
You have written the values as D = 10 and PI = 3.14159
D = 10, A= 78.5398 and C= 31.4159.
Also note that in the above program we have used endl ; (end line) instead of std::endl;
because we have used the directive using namespace std; in the beginning of the program.

2.7 FORMATTING THE OUTPUT

Often we need to format the output. There are many functions for the same but at present we
take only a few that are provided in <iostream>. However, more details are discussed in Chapter
19. Here we discuss only the following.
(i) Putting spaces between items of output, i.e. use of “ ” or tab “\t”
(ii) Controlling width of output.
(iii) Control of precision for float and double numbers.
(iv) To put the floating decimal point number in scientific notation.
(v) Use of fill() function to fill empty spaces.
(vi) Codes for right and left justification.
First we take the case of providing spaces. Even if you write the output statement in a
fragmented manner as illustrated in Program 2.7, they would be put together by operator <<
unless spaces are specified. One method of providing space is by enclosing the blank spaces in
double quotes like “ ”. Because any statement or blank spaces enclosed between double quotes
is sent to output stream as it is written. Alternatively, space character “\t” called tab may be used
to provide a preset space. It has to be enclosed in double quotes like “\t” and you may use single
“\t” or multiple tabs as in “\t\t\t”, etc. The number of spaces for each “\t” may also be set.

PROGRAM 2.7 – Multiple use of operator << , the tab “\t” and “ \n ” in output statement.

#include <iostream>
using namespace std;
int main()
{
cout<<“Hello”<<“ Nat”; // Natasha is broken into pieces
cout<<“as”<<“ha,”<<“\n\tGood”<<“ Morning!\n”; // use of “\n” and “\t”
cout<<“When are you going to market?”<<endl;
return 0;
}
The output of the above program is given below. You will notice that even though the words
have been fractured into pieces in cout statements, but in output these are placed together one
after another in the same order.
❖ 40 ❖ Programming with C++
Hello Natasha,
Good Morning!
When are you going to market?
The above program also illustrates the use of tab “\t” to provide space. The second line of
output is printed with a margin because “\t” has been put before Good and after new line
character “\n” in the program,( see “\n\tGood” in the program).
The following program shows that assignment can also be done by putting the value to be
assigned in parentheses following the variable identifier. Moreover, the values are assigned in
scientific notation.

PROGRAM 2.8 – Illustrates scientific notation and assigning the values.

#include <iostream>
using namespace std;
void main()
{
double C (4.56E9); // Number expressed in scientific notation
// the value is assigned by putting it in brackets
double D = 6.54e9;
double E = 675.987654;
char Kh(‘B’); // assignment by putting the value in brackets.

cout <<“C = “ << C <<“ “<<“, D = “ << D<<endl;

cout <<“ C+D = “<< C+D<<endl;

char ch = ‘A’;
cout <<“Kh = “<<Kh <<“ “<<“ch = “<<ch <<endl;
cout <<scientific << E<<endl;
// asking for output of E in scientific notation
}

The expected output is given below.

C = 4.56e+009 , D = 6.54e+009
C+D = 1.11e+010
Kh = B ch = A
6.759877e+002

The above program illustrates that we can also assign a value to variable by putting the value
in parentheses immediately following the variable name. Another example is given below.
double C(4.56E9);
 Structure of a C ++ Program ❖ 41❖
This is equivalent to the following.
double C = 4.56 E9;
Also the code kh(‘B’); assigns the value ‘B’ to char kh.
We may also ask for output in scientific notation. This is illustrated in the last output
statement in the program in which the code <<scientific is added in the output statement, as a
result of which the number 675.987654 has been displayed as 6.759877e +002 in the output.

FUNCTION WIDTH ()
The function width() is associated with the object cout and is coded as below.
cout.width(unsigned int n);
Here n, a positive number, specifies the spaces desired in output. The following program
illustrates the use of function width()and left and right justification. The default setting is right
justification. If the output characters are less than n spaces, the remaining will be blank spaces.
PROGRAM 2.9 – Illustration of function width() and right and left justification.

#include <iostream>
using namespace std;
int main()
{ int x ;
double A ,y ;
cout<<“Write one int and one double number: ”;
cin >> x >>y;
cout.width(20);
cout.setf(ios::right); // right justified
cout<<“x = “<<x<<endl;
cout.setf(ios::left); // left justified
cout<<“x = “<< x<<“\n”;
cout<<“y = “<<y<<“\n”;
A = x+y;
cout<<“Value of A = “ <<A<<“\n”;
return 0;
}
The expected output is given below.
Write one int and one double number: 4000000 5.2E8
x = 4000000
x = 4000000
y = 5.2e+008
Value of A = 5.24e+008

As already mentioned the right justification is the default justification, if left justification is
desired it has to be specified. The ios is the base input /output class. The detailed discussed of
these is given in Chapter 19 in which ios class is discussed. Since we have not dealt with classes
❖ 42 ❖ Programming with C++
as yet, we take it as a code. The second observation of the output is that if input is in scientific
notation output is also in the same notation.
FUNCTION PRECISION ()
In the next example we take up the use of precision() function along with width()
function. Code for precision() is similar to that for width()function. It is illustrated below.
cout.precision(unsigned int m);
where m represents the number of digits desired in the output. The function precision ()
is linked with object cout by dot (.) operator. Function precision () is used only for
floating decimal point numbers, i.e. for float and double.

PROGRAM 2.10 – Illustrates application of function precision().

#include <iostream>
using namespace std;
int main()
{
int A ;
double PI ,B ;
A = 4567543;

PI = 3.141592653589793238;
B = 245.7654329832;
//If precision is not specified, default precision is 6
cout <<“PI= “<< PI <<“\tA = “<<A<<“\tB = “<<B<<endl;
cout.precision(4);

cout <<“PI= “<< PI <<“\tA = “<<A<<“\tB = “<<B<<endl;

cout <<“PI= “<< PI <<“\tA = “<<A<<“\tB = “<<B<<endl;

cout.precision(8);
cout <<“PI= “<< PI <<“\tA = “<<A<<“\tB = “<<B<<endl;
return 0;
}

The output is given below.

PI= 3.14159 A = 4567543 B = 245.765
PI= 3.142 A = 4567543 B = 245.8
PI= 3.142 A = 4567543 B = 245.8
PI= 3.1415927 A = 4567543 B = 245.76543
See the result carefully. The first line of output is the default output without any specification
in the program. 6 digits are printed for decimal variable while integer is printed in full. For the
 Structure of a C ++ Program ❖ 43❖
second line of output we specified precision(4). So only 4 digits have been printed for the
floating decimal point numbers, while the integer A has been printed full. For the last line of
output we increased the precision to 8. For the numbers with decimal points 8 digits are displayed
while the integer number is written full. The precision setting continues till a new setting is
encountered. In the next example we take the case of insufficient width setting. When the width
setting is less than that required for output the width setting is neglected.

PROGRAM 2.11 – Illustrates the case of insufficient width setting.

#include <iostream>
using namespace std;
int main()
{ int A = 867564;
double p = 4.532678;
double m = p/5;

cout.width(12);
cout <<A <<“\n”<< p<<“\n”<< m << endl;

cout.precision(4);
cout.setf(ios::left);
cout <<“A = “<<A<<“\t”<< “p = “<<p<<“\t”<< “m = “<<m << endl;

cout.precision(20);cout.width(10);
// width setting is less than precision setting.
cout <<p<<“\t” << m << endl;
return 0;
}

The output is given below.

867564
4.53268
0.906536
A = 867564 p = 4.533 m = 0.9065
4.5326779999999998 0.9065356
Compare the outputs with the specifications given in the cout statements. The effect of width()
setting remains only up to the first line of output. In the first line of the output A is printed in
12 spaces on the right side. It has 6 digits and there are 6 blank spaces on the left of it. The second
and third lines of output are not affected. Also for the second and third line of output the default
precision setting is 6 so 6 digits are displayed in output. For the 4th line of output the precision
setting is 4 so only 4 digits are displayed in case of p and for m. Integer numbers are not affected
by precision setting. For the next line of output, precision is set at 20 while the width setting is
10. Therefore the width setting has been neglected. While the width setting affects only the next
line of output, the precision setting remains till it encounters a new setting.
❖ 44 ❖ Programming with C++
FUNCTION FILL ( )

If the width setting is more than the number of characters in the output, there would be some
empty spaces which may be filled with a character such as dash(-) or some other character by
using fill () function. The character desired to fill with is put in the argument. It has to be enclosed
by single quotes. Thus for filling with character dash(-) we write the code
cout.fill(‘–’);

See the following program for illustration.

PROGRAM 2.12 – Illustrates application of fill() function.

#include <iostream>
using namespace std;
int main()
{
cout.width(15); //hello will be written in 15 spaces
cout.fill(‘-’); //fill the empty spaces with dash(-)
cout<<“Hello,”<<endl;
cout.width(40); //the output be written in 40 spaces,
cout.fill(‘*’); //fill the empty spaces with ‘*’
cout<<“Welcome to”<<“ “<<“programming in C++!”<<endl;
cout.width(36);
cout.fill(‘$’); // fill empty spaces by ‘$’
cout<<“Are you learning C++ ?”<<endl;
return 0;
}
The expected output is given below.
---------Hello,
******************************Welcome to programming in C++!
$$$$$$$$$$$$$$Are you learning C++ ?

The output shows that first line of output is in 15 spaces and 9 spaces have been filled with
dash. In the next line of output, the 40 spaces are up to end of ‘Welcome to’. Moreover no filling
has taken place in the blank spaces in between the statement.
The cout.fill() function persists beyond the immediate next line of output statement
till next fill() function is encountered. This is illustrated in the following program. For
undoing the previous setting, include an empty fill setting as illustrated below. For more details
see Chapter 19.
cout.fill(‘ ’);
 Structure of a C ++ Program ❖ 45❖
PROGRAM 2.13 – Illustrates fill() function along with width() and precision().

#include <iostream>
using namespace std;
int main()
{
int x =205;
double PI=3.141592653589793, y=4564.2318765;

cout.fill(‘-’);
cout.width (6);
cout<<x<<endl;

cout.precision(8); cout.width(16);
cout<<PI<<endl;
cout.precision(6);
cout.fill(‘ ’);
cout.width(10);
cout<<y<<endl;

return 0;
}

The expected output is as below.

---205
-------3.1415927
4564.23

In the above program the empty spaces are desired to be filled with dash (‘-’) in the output
for x. Therefore the function cout.fill(‘-’); has been added before the output statement
for x. In the next output statement for PI the function cout.fill() has been omitted but
we find that filling by dash continues. Therefore, for removing the fill() function setting an empty
fill function cout.fill(‘’); has been added before the output statement for y, so no fill
symbol appears in output of y. For more details see Chapter 19.

2.8 FUNCTION cin.getline() v/s cin

For entering a string of characters as we come across in names, messages etc., we can make use
of getline(). It is coded as below.
char message [20]= “Good morning!”
cin.getline (message, int n, ch) ;
Here message is a string of characters. It can have 20 characters including the Null character
(‘\0’) which marks the end of string (see Chapter on C-strings for more details on strings).
❖ 46 ❖ Programming with C++

Of the arguments of getline(), ‘message’ is the name of string, the int n (a positive number)
is the number of characters to be read and ch is the name of character on encountering of which
it would stop reading. This is illustrated in the following program. In the program the user is
asked to enter a name. The maximum number of characters to be read is specified as 30. The
limiting character ch is specified as below.
char ch = ‘L’;

Whenever a value is assigned in form of a character, it should be enclosed in single quotes

as shown above.

PROGRAM 2.14 – Illustrates the application of function getline().

#include <iostream>
using namespace std;
int main()
{
char Name1 [30] ;
char ch = ‘L’;
cout<< “Enter a short name :”;
cin.getline(Name1, 30, ch);
cout<<“\nName1 = ” << Name1;
return 0;
}

The expected output is as given below.

Enter a short name :Tata Motors Ltd

Name1 = Tata Motors

The name entered was Tata Motors Ltd. The output shows that on encountering the letter
‘L’ the reading stopped. So output is only Tata Motors.
The following program illustrates the difference between cin and cin.getline (). The
function getline() reads the whole line including the blank spaces while cin does not read
blank spaces. It takes a blank space as end of the value.
PROGRAM 2.15 – Illustrates difference between cin.getline() and cin.

#include <iostream>
using namespace std;
int main()
{
char word [30];
cout<<“Write two small words : ”;
cin.getline(word,30);
 Structure of a C ++ Program ❖ 47❖
cout<<“You have written ”<< word<<“\n”;
cout<<“Write two small words : ”;
cin>> word;

cout<<“You have written ”<< word<<“\n”;

return 0;
}

The expected output is given below.

Write two small words : TATA MOTORS
You have written TATA MOTORS
Write two small words : Tata Motors
You have written Tata

On running the program the following line appears on the monitor

Write two small words :

So two words TATA MOTORS were typed and entered. The function cin.getline ()
reads the two words and the output is given in the next line of output. Again another line given
below appears on screen.
Write two small words :

This time the words entered are Tata Motors. These are read by cin. From the output you
see that only Tata is printed. This is because cin stops reading when a blank space is encountered.
While in case of getline() it reads blank spaces as well as characters.

EXERCISES
1. What are the program tokens in a simple C++ program?
2. What do you understand by expression ‘ # include <iostream>’ in a program ?
3. How do you declare an integer variable, a variable with value in floating decimal point number
and a character variable in a C++ program?
4. What for fill() function is used ?
5. Is it mandatory to write return 0 ; as last line in all the C++ programs?
6. How would you include comments in the listing of a program?
7. What are header files?
8. What are void functions?
9. What is C++ Standard Library?
10. What do you understand by terms ‘namespace’ and ‘namespace std’ and the directive ‘using
namespace std ; ’?
11. What do the following stand for? What are they called?
❖ 48 ❖ Programming with C++

(i) “\t”
(ii) “\n”
(iii) /* */
(iv) //
12. Make a program to illustrate the effect of precision setting when the argument of the function
precision is less than the number of digits before the decimal point.
Answer:

PROGRAM 2.16 – Illustrates precision () and width() functions

#include <iostream>
using namespace std;
void main()
{
double PI = 3.14159265358;
double B = 67543687.897;
cout.precision(4);
cout<< “B = ” << B <<“ PI = ”<<PI<<endl;
cout.setf(ios::fixed); //output in normal notation.
cout.precision(5);
cout<<“B = ” <<B <<endl;
}

The output shown below shows that in scientific notation the output consists of 4 digits.
For output in fixed notation the precision setting has given 5 digits after decimal point.
B = 6.754e+007 PI = 3.142
B = 67543687.89700
13. Select the correct type of variable for the following quantities.
(i) Temperature of hot water in degrees centigrade.
(ii) Volume of a bucket.
(iii) Number of machines in a workshop.
(iv) For writing values like ‘A’, ‘B’ etc. in a program
14. How would you write input and output statements when the directive ‘using namespace std;’
is not used?
15. What do you understand by function precision () ?
16. How would you unset the already set fill() function?
17. Find errors in following declarations and write the correct versions.
(i) Int n ;
(ii) int 8 = n;
(iii) char ch = a ;
(iv) Char alpha = A;
 Structure of a C ++ Program ❖ 49❖
(v) Double M = 6.78;
18. What for we use the function width()?
19. Which of the following statements are true?
(i) Every C++ program must have one main () function.
(ii) Statements in a C++ program must end with a semi-colon (;).
(iii) In C++ all the variables must be declared in the beginning of the program.
(iv) A C++ program always ends with return 0 ; .
20. Find errors in the following program. Rewrite a correct version of it.
include <iostream>
int main ()
int n , m ;
{n = m + 4;
m = 5.5;
cout << m << “ “n<< endl;
cout >> m*m <<endl;
}
21. Write a program to make character ‘D’ with character (‘*’) .
22. Make program for finding area of rectangle with sides in floating point numbers.
23. Make a program to make filled triangle with apex at the top by using (‘*’) for filling.
Answer: The following program is made to write different number of asterisk(*) in successive
lines to make a solid triangle as illustrated in the output below.

PROGRAM 2.17 – Illustrates program for making a figure – solid triangle

#include <iostream>
using namespace std;
int main()
{cout<<“ *\n ***\n *****\n *******\ n
*********\n***********”<< endl;
return 0 ;
}

The output is given below

*
***
*****
*******
*********
***********
24. Make a program for making filled right angle triangle with base parallel to horizontal plane.
Answer:
❖ 50 ❖ Programming with C++
PROGRAM 2.18 – Illustrates program for making a solid right angle triangle.

#include <iostream>
using namespace std;
int main()
{ cout<<“\t *\n\t **\n\t ***\n\t ****\n”;
cout<<“\t *****\n\t******”<< endl;
return 0 ;
}

The output of the program is given below.

*
**
***
****
*****
******
25. Make a program to make a figure of letter H with the symbol (*) .
Answer to Q.25

PROGRAM 2.19 – Illustrates program for making a figure like the character H

#include <iostream>
using namespace std;

void main()

{ cout<<“\t* *\n\t* *\n\t* *\n\t********\n”;

cout<<“\t* *\n\t* *\n\t* *\n” ;
}

The output is given below.

* *
* *
* *
********
* *
* *
* *
26. Make a program to print the message “Hello, Welcome to Programming in C ++”. The print
statement in the program be written in two lines with the message broken as “He” , “ll”, “o”,
welcome” in first line and “ to programming in C++ “ in second line.
Answer:
 Structure of a C ++ Program ❖ 51❖
PROGRAM 2.20 – An exercise in output statement
#include <iostream>
using namespace std;
int main()
{
cout<<“He”<<“ll”<<“o,”<<“Welcome”;

cout<<“ to programming in C++!\n”;

return 0;
}
The output is given below.
Hello, Welcome to programming in C++!
27. Make a program to print T and S with character ‘*’ as done for H in Q.25.
28. Make a program to print the + sign with the help of * symbol.
29. Write the expected output of following program.

PROGRAM 2.21 – An exercise to illustrate use of “\n” and “\t”

#include <iostream>
//use of tab, namespace, new line character.
using namespace std;
int main()
{
cout<<“\tHello,\n\tWelcome to\n programming in C++!\n”;
return 0;
}
Answer:
The expected output is as below.
Hello,
Welcome to
programming in C++!
30. Write the expected output of following program.

PROGRAM 2.22 – Exercise for finding output of a program

#include <iostream>
int main()
{
std::cout<<“Hello,\a\’Are you coming to C++ class? \’\n”;
std::cout<<“Hello,\\Are you coming to C++ class?\\\n”;
std::cout<<“ Are you coming to C++ class?\rMohan\n”;
❖ 52 ❖ Programming with C++
std::cout<<“Hello\tAre you coming to C++ class?\n”;
return 0;
}

Answer:
The expected output is given below.
Hello,’Are you coming to C++ class? ‘
Hello,\Are you coming to C++ class?\
Mohan Are you coming to C++ class?
Hello Are you comming to C++ class?
31. Identify errors in the following program. Correct them and run the program.

// Program for correction.

#include <iostream>;
using namespace STD;
int main();
{
int x =205;
double PI=3.141592653589793, y=4564.2318765;

cout.fill(‘-’);
cout.width (6)
cout<<x<<endl;

cout.precision(10); cout.width(8);
cout<<PI<<endl
cout.precision(6);
cout.fill(‘ ’);
coutwidth(10)
cout<<y<<endl;

return 0;}

❍❍❍
CHAPTER

++

3.1 FUNDAMENTAL DATA TYPES

In C++ data is differentiated into different categories or types such as integers or whole numbers
(type is int), decimal point numbers (type is float or double), characters (type is char), etc.
However, C++ also allows user defined types such as class objects or structures for which the
name of the class or the structure is their type (see Chapter 11). The data types which are implicitly
defined in C++ are the fundamental data types which form the subject matter of this chapter.
We all know that computers work on numbers expressed in binary. Anything that is entered
into a computer whether these are numbers which we use in our daily calculations or these are
instructions and operators such as +, –, / ,etc., or are alphabets such as A, b, C, etc., all are stored
in the form of binary numbers which are sequences of 0 and 1. For instance, when we type text
on keyboard connected to computer the letters are stored in computer memory in the form of
binary numbers according to ASCII code which is followed universally (see Appendix A). ASCII
(pronounced as ‘as-key’) stands for American Standard Code for Information Interchange. In this
code, every letter or symbol has a value. For instance, character ‘A’ has value 65 and is stored as
1000001 in computer memory, similarly character ‘B’ is equivalent to 66 and computer stores
it as 1000010 and so on. Also when we retrieve the same data on the computer monitor or
printer it should display or print A for ‘A’ and B for ‘B’ and not 65 and 66. That implies that
the type of data that is entered or retrieved from computer must be specified.
As per ASCII code the characters and symbols have values which range from 0 to 127 and
thus can be stored in one byte of memory. A number like 6574594 or like 123452.9876754326
would require much more memory space. Also whole numbers and numbers with decimal point
are stored differently and require different memory spaces. When a variable is declared by putting
its type followed by its name (or identifier), the compiler allocates a block of memory space for
storing the value of the variable. The size of the memory block depends on the type of the
variable. For every type of variable a different but a limited space (in number of bytes) is allocated
by the compiler, the size of which depends on the system hardware, operating system and the
compiler.
By declaring the type of a variable, the type of data that the variable can have also gets known.
For example, if the type of a variable is int, the variable is an integer or whole number. The
values that may be assigned to it should also be integers or whole numbers. Moreover, with each
❖ 54 ❖ Programming with C++

type a particular set of operations are associated. Thus by specifying the type we, in fact, refer to
a class of variables on which a particular set of operations may be performed. Besides integers,
decimal point numbers and characters, there are other types in C++ such as Boolean type,
enumeration type, etc. Table 3.1 below lists the different fundamental types in C++.
Table 3.1 – The fundamental types in C++ and the codes

Type Code
(A) Integral type – These comprise the following types.
(i) Boolean type bool
(ii) Enumeration type enum
(iii) Character type which includes
(a) Characters char
(b) Unsigned characters unsigned char
(c) Wide character wchar_t
(iv) Integers – These comprise
(a) Short integers short
(b) Integers int
(c) Long integers long
(d) Unsigned short integers unsigned short
(e) Unsigned integers unsigned int
(f) Unsigned long integers unsigned long
(v) Floating decimal point types – These comprise
(a) Normal precision or single precision float
(b) High precision or double precision double
(c) High precision or double precision long double

INTEGERS
Integers are associated with every aspect of our life. A simple example is that of counting, such
as number of members in a family, number of passengers in a bus, number of gun shots fired
or number of events, etc. In all such cases whole numbers are used. Because, there cannot be a
half passenger in bus or one quarter member in a family or one fourth shot fired. The whole
numbers whether they are positive or negative are called integers. A whole number may be short
such as number of members in a family or very long number such as number of seconds in a
century. Naturally they require different sizes of computer memory for storage. The program
must allocate sufficient memory to store the short as well as long number. If the number is bigger
than the memory allocated then a part of number may get truncated or reduced in size which
will lead to errors. Also if memory allocated is too big it would be wasteful use of memory space.
Therefore, whole numbers are differentiated according to their size. Thus there are three types
of whole numbers as distinguished by their size. They are called short, integer and long and the
codes for the three are written as short, int and long.
On most of the 32 bit computer systems short is stored in 2 bytes. Two bytes comprise 16
bits. Out of the 16 bits the leftmost bit is reserved for sign (+ or –), hence only 15 bits are
 Fundamental Data Types in C++ ❖ 55❖
available for storing the number. Therefore, the range of numbers that may be stored as short is
from – 215 to + (215 – 1), that is from – 32768 to + 32767 (see Chapter 1 for details).
On a similar system, int is allocated 4 bytes (32 bits). In this case also the leftmost bit is
reserved for sign ( + or –). Therefore, the range of numbers that can be stored varies from – 231
to + (231 – 1) that is from – 2147483648 to + 2147483647. The long is also allocated 4 bytes,
therefore, its range is also the same as for int.

UNSIGNED INTEGERS
There are many applications where only positive integers are permitted. To illustrate this point
we take the case of computer memory which consists of bits. These are grouped into bundles
of 8 each called bytes. In RAM bytes are numbered in a sequence. A number in this sequence is
the address of a byte. This number cannot be negative. It is just like that your house number
cannot be negative. These are called unsigned numbers. The signed numbers can be positive or
negative, but unsigned numbers are always positive. The unsigned numbers are also short
unsigned numbers, integer unsigned numbers and long unsigned numbers. These types are
respectively coded as unsigned short, unsigned int and unsigned long. On a 32-bit system
unsigned short is allocated 2 bytes, but no bit is needed for sign, therefore, the range of unsigned
short is from 0 to 65535. Similarly the range of unsigned int on most systems ranges from 0
to (232 – 1), i.e. from 0 to 4294967295.

FLOATING DECIMAL POINT NUMBERS

There are quantities which cannot be expressed as whole numbers. For example, if you take your
weight, it is quite unlikely that it would be an exact number of kilograms. It may be to say 55.25
kilogram. The number 55.25 can also be written as 221/4, i.e. a fraction. Therefore, such numbers
are also called fractions or floating point numbers. The term floating point has originated from
the method the number is stored in computer. Say the number 75.25 is to be stored. In binary
it is written as 1001011.01. The number is stored as .100101101 × 27. On a 32 bit system the
number 1001011 01, is stored on the 23 bit segment of the 32 bits while the power 7 or 111
in binary, is floated, i.e. it is stored separately on the remaining 8 bit segment.
The leftmost bit (32nd bit) is kept for sign (+ or –). In scientific notation also we often float
the decimal point. The number 7685.43 may be written as 7.68543 × 103. Therefore, in
scientific notation the number 7685.43 is expressed as 7.68543 e+3 or 7.68543 E+3 in C++.
So such numbers are called floating point numbers. These numbers may also be +ve or negative.
Both the integers and fractions when grouped together are called rational numbers.
Now consider a number like π which is the ratio of circumference of a circle to its diameter.
This number cannot be expressed exactly as ratio of two whole numbers or a fraction. It can only
be approximated to a ratio. For instance, we often use the approximate value of π as 22/7. In
the same category are the numbers like √2, √3 or √5, etc. All such numbers are called irrational
numbers. They are also stored just like floating point numbers.
As far as computer program is concerned we differentiate between whole numbers or integers
and numbers with floating decimal point because they are stored differently in the computer
memory. We also discussed that whole numbers are differentiated according to size as well.
❖ 56 ❖ Programming with C++

Similarly, the floating point numbers are differentiated according to the precision required, i.e.
the number of digits after the decimal point. They are respectively called and coded as float,
double and long double. On most 32 bit systems a float has seven significant digits and is allocated
4 bytes of memory while a double has 15 significant digits, nearly double the precision of float.
It may get 8, 10, 12 or 16 bytes for storage on different systems. The memory spaces generally
allocated for a float, double and long double are given in Table 3.3.

CHARACTERS
The alphabets along with some other symbols are called characters. In C++ characters are stored
according to ASCII code (See Appendix A). For example, a few of the values are listed below.
Character Value as per ASCII Character Value as per ASCII
A 65 a 97
B 66 b 98
—— —— —— ——
0 (digit) 48 1 (digit) 49
5 (digit) 53 9 (digit) 57
According to this code the numerical values of alphabets and other symbols range from 0
to 127 and in the extended IBM list up to 255, which can be written in 8 bits or one byte.
Therefore, each character is allocated one byte for its storage.

THE ENUMERATION TYPE

As described in the introduction, fundamental types are defined in C++, besides a programmer
may create his/her own data types through classes. The enumeration type may also be used to
define user’s own data type. The enumeration type is an integral data type. Its syntax is as follows.
enum typename { enumerator_list_of_objects};
The syntax is illustrated below with examples.
enum Day { Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday};
In the above case, Monday would have a value 0, Tuesday would have value 1, Wednesday
2 and so on. In the above declaration if you put Monday = 1 as illustrate below then, Tuesday
would have value 2, Wednesday 3 and so on.
enum Day { Monday =1, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday};
The enum may also be used to define a number of integral constants as given below.
enum Data { A = 100, B = 100, C = 400, D = 500 , E = 600};
or
enum Marks { correct =2, wrong = – 1, not_attempted = 0};

BOOLEAN TYPE
Yet another type is the Boolean type integer which is coded as bool. When we examine a logical
 Fundamental Data Types in C++ ❖ 57❖
statement it may be either true or false. We may associate a number say 1 for true and 0 for
false. For example, let us have a conditional statement such as, if (A > B) in a program. Now if
this condition is actually true, 1 will be sent to the system. If it is false, 0 will be sent to the
system. In this way the system will know whether the condition is true or false.

3.2 DECLARATION OF A VARIABLE

Declaration of variables was briefly discussed in Chapter 2, more details are given below. A
variable may be declared as below.
type identifier ;
Here type is the data type such as int, float or double, char, etc., as listed in Table 3.3. A
variable must have a name with which it is identified. It is just like that you are identified by
your name. The type must be written in lower case. For example an integer with name Age is
declared as below
int Age;
Note that the line ends with a semicolon (;). Any variable before it is used in the program
should be initialized as well, i.e. a value should be assigned to it. In C++ it is not necessary to
declare and initialize all the variables right in the beginning. It may be done anywhere in the
program but before it is actually used in the program. Initialization is done by using assignment
operator (=) as illustrated below.
int Age;
Age = 22;
The above two lines may be combined as given below.
type identifier = initial value;
int Age = 22;
double PI = 3.1415926535;
The computer output for both float and double will be according to the default precision
set in the computer which is generally 6 digits. However, you can change it by specifying another
precision value. Code for setting precision has already been discussed in Chapter 2.
A variable of type character is also declared as below.
char identifier ;
For example a variable with name ch and value ‘A’ is declared and assigned as below.
char ch = ‘A’ ;
With this definition of ch, the compiler would allocate one byte for storing the value of ch.
It may also be declared as a string as given below.
char ch[] = “A”;
In this case two bytes are allocated by compiler for ch[ ], i.e., one for ‘A’ and one for NULL
character (‘\0’) which is appended by the system and marks the end of string ( see the output
of Program 3.16).
❖ 58 ❖ Programming with C++

Below is another example of declaration of a variable which has value as a string of characters.
In all such cases the value is enclosed in double quotes and number of characters in the string
+1 are put in square brackets.
char River[6] = “Ganga”;
Here River is the name of a char variable and in the above definition its value is “Ganga”.
In the above declaration the number of elements of the string “Ganga” are 5 plus the Null
character, so total is 6. When the declaration and initialization are together we need not put the
number of elements. The compiler can count the number of elements. Thus we could have also
written as below.
char River [ ] = “Ganga” ;
Consider the declarations of following variables, i.e. Digits and Plus are declared and values
‘5’ and ‘+’ are assigned respectively.
char Digit = ‘5’ ;
char Plus = ‘+’ ;
Here Digit is the name of a variable of type char and character ‘5’ is assigned to it as its value.
According to ASCII code, character ‘5’ has a value 53. Similarly ‘+’ is assigned to variable Plus.
The character ‘5’ is not the usual 5 used in calculations. Similarly here ‘+’ is not the usual operator
+. Here it is a character. The characters as per ASCII code have values from 0 to 127 (IBM
extended list is up to 255) and therefore, are allocated one byte for storing one character. The
wchar_t, also called as wide character type, cannot be stored in one byte. Two bytes are allocated
for its each character. It belongs to Unicode character set which is an international standard
character set and is also supported by C++. This character set includes characters of different
languages of world such as Latin, Greek, Bengali, Devnagari, Gurmukhi, geometric shapes,
mathematical symbols, technical symbols, etc. The interested readers are advised to refer to
reference “The Unicode Standard, Version 2.0 by Unicode Consortium” - Addison- Wesley, 1996.

3.3 CHOOSING AN IDENTIFIER OR NAME FOR A VARIABLE

Computer identifies a variable by its name just as you are identified by your name. When a
variable is declared the compiler allocates a block of memory for placing the value of the variable
in it. In fact, for the computer that allocated block of memory is the variable and you can get
to the value stored in this memory block by calling the name of the variable. The memory sizes
allocated for different types may vary on different computers depending upon the hardware, the
operating system and the compiler used. Table 3.3 gives the different types of variables and the
sizes of memory blocks (number of bytes) generally allocated for different data-types on most
computers. You may determine the same for your computer by running the Program 3.2, on
your computer. The names of variables or identifiers should be carefully selected, because a valid
identifier should comply with the following points.
1. It can be a sequence of one or more characters including underscore symbol ( _ ).
Example - Barlength or Bar_length are valid names.
 Fundamental Data Types in C++ ❖ 59❖
2. There should be no white (blank) spaces in an identifier. It should be a single word.
Bar length is invalid. Box volume is invalid while Boxvolume or Box_volume are valid.
3. There should be no marked character in the identifier.
4. There is no limit on length of name in C++. However, in C it is limited to 32
characters. Better have a check on the compiler you are using.
5. Identifier should never start with a digit. It should always start with an alphabet or
underline symbol, however, digits can be part of identifier. 2A is invalid while A2 and
_2A are valid.
6. It should not have any other symbol except alphabets, digits and underscore symbol.
Example- Newdelhi_2 is valid but Newdelhi-2 or Newdelhi(2) are invalid identifiers.
7. It must not be a keyword of C++. A list of keywords is given in Table 3.2. Thus double
or int are invalid identifiers while Double or Int or INT are valid identifiers.
8. Besides the keywords the specific reserved words for the compiler should not be used
as identifiers.
9. The alternative representations of some operators of C++ also should not be used
as identifiers. These 11 words are given below.
and, and_eq, bitand, bitor, compl, not, not_eq, or, or_eq, xor, xor-eq
If one or more characters in the name is in upper case it is a safe bet against a keyword being
used as name of a variable. A list of keywords is given in Table 3.2 below.

3.4 KEYWORDS

There are 63 keywords in C++. Out of these 32 are common with C. The keywords in C++
are listed in the table below. The common keywords are shown bold.
Table 3.2A – Keywords in C++
asm auto bool break case catch
char class const const_cast continue default
delete do double dynamic_cast else enum
explicit export extern false float for
friend go to if inline int long
mutable namespace new operator private protected
public register reinterpret_cast return short signed
sizeof static static_cast struct switch template
this throw true try typedef typeid
typename union unsigned using virtual void
volatile wchar_t while

Table 3.2B – The 11 alternative representations of some C++ operators

and and_eq bitand bitor compl not

not_eq or or_eq xor xor-eq
❖ 60 ❖ Programming with C++
PROGRAM 3.1 – Illustrates declaration of different types of variables.

#include <iostream>
using namespace std;
int main ()
{ int Dia = 10 ; // variable Dia stand for diameter.
double PI = 3.14159265358979323;
float pi = 3.14159265;
double Area_circle = PI* Dia*Dia / 4;
cout<<“PI = “<<PI<<“\t pi = “<<pi<<endl;
char ch = ‘S’; // ch is the name of a variable with value ‘S’
cout << “ch = “ << ch <<endl;
cout<< “Area_circle = “ << Area_circle<< endl;
int m, n, p; // m, n, p are names of variables of type int
m = 4; // 4 is value of m
n = 3;
p = m*m + n*n ;
cout <<“p = “<<p << endl;
return 0;
}

The expected output of the program is as below. The output for both float and double is
according to the precision setting in the computer. In present case it is 6, so output is in 6 digits.
PI = 3.14159 pi = 3.14159
ch = S
Area_circle = 78.5398
p = 25

The following table lists most of the fundamental types and the memory allocated in general.
Table 3.3 – The different fundamental types and memory allocated for different types

S.No. Variable Code for type Examples of Memory(bytes)

type declaration allocated in
general
1. character char char A, char ch; 1

2. signed character signed char signed char ch; 1

3. boolean bool bool M ; 1

4. wide character wchar_t wchar_t ch; 2

5. integer int int x; 4

6. short integer short short n; 2

7. long integer long long y; 4

Contd...
 Fundamental Data Types in C++ ❖ 61❖

8. unsigned short unsigned short unsigned short M; 2

9. unsigned integer unsigned int unsigned int N; 4

10. unsigned long unsigned long unsigned long X; 4

11. float float float Weight ; 4

12. double double double Area ; 8

13. long double long double long double Area; 8

3.5 SIZE OF FUNDAMENTAL DATA TYPES

The function sizeof() returns the number of bytes allocated for its argument. The memory
allocation for different types of data may be different in your computer because it depends upon
the hardware and software being used. The following program makes use of the function
sizeof() to determine the size of different types. It also includes wchar_t characters.

PROGRAM 3.2– Illustrates the application of function sizeof().

#include <iostream>
using namespace std;
//The program finds the memory allocated for different data types
int main()
{cout<<“Number of bytes reserved for following types.”<<endl;
cout<<“ For character : \t” <<sizeof(char)<<“\n”;
cout<<“ For signed character : \t”<<sizeof(signed char)<<“\n”;
cout<<“ For wide character : \t”<<sizeof (wchar_t)<<“\n”;
cout<<“ For integer : \t”<<sizeof(int)<<“\n”;
cout<<“ For short : \t”<<sizeof(short)<<“\n”;
cout<<“ For long : \t”<<sizeof(long)<<“\n”;
cout<<“ For float : \t”<<sizeof(float)<<“\n”;
cout<<“ For double : \t”<<sizeof(double)<<“\n”;
cout<<“ For long double : \t”<<sizeof(long double)<<“\n”;
cout<<“ For unsigned short : \t”<<sizeof(short unsigned)<<“\n”;
cout<<“ For unsigned integer : \t”<<sizeof(int unsigned)<<“\n”;
cout<<“ For unsigned long : \t”<<sizeof(long unsigned)<<“\n”;
wchar_t Wch = L’H’; // wide characters are preceded by L.
cout << “ The size of Wch : \t”<<sizeof (Wch)<< “\n”;
return 0;
}

The expected output is given below,

Number of bytes reserved for following types.
❖ 62 ❖ Programming with C++
For character : 1
For signed character : 1
For wide character : 2
For integer : 4
For short : 2
For long : 4
For float : 4
For double : 8
For long double : 8
For unsigned short : 2
For unsigned integer : 4
For unsigned long : 4
The size of Wch : 2 //wide character takes 2 bytes

The following program illustrates the declaration of char variables and shows that some
arithmetic operations may also be performed with them just as they are done on integral digits.
You will also note that a digit such as 9 will have different value when it is used as a character (‘9’).

PROGRAM 3.3 – Illustrates use of variables of type char.

#include <iostream>
using namespace std;
int main()
{ char ch1,ch2,ch3, ch4, ch5 ,ch6,ch7,ch8;
// ch1 to ch8 are simply names of variable. Their type is char

ch1= ‘A’; //The value of character A in ASCII code is 65.

ch2 = ‘z’; // Value of ‘z’ in ASCII code is 122.
ch3 = ch2;
ch4 = ch2 - ‘9’; // here ‘9’ is character and not number 9.
// Value of character‘9’in ASCII code is 57.Thus 122-57 = 65=’A’.

ch5 = ‘)’*2;
//value of character’)’ in ASCII code is 41, so 41 × 2 =82= character R
ch6 = ch1 + 10; //’A’=65,so 65 +10 = 85 =’K’
ch7 = ch2 % ch1; /* This operation gives remainder of
122/65 which is 57= ‘9’ */
ch8 = ‘9’*2; // ‘9’ = 57 on ASCII code so 57 × 2 = 114 = ‘r’
cout<<“ ch3 = “<<ch3 <<“\t ch4 = “ <<ch4 <<“\tch5 = “<<ch5<<endl;
cout<<“ ch6 = “<<ch6<< “\tch7 = “<<ch7<<“\tch8 = “<<ch8<<endl;
return 0;
}
 Fundamental Data Types in C++ ❖ 63❖
The expected output is given below.
ch3 = z ch4 = A ch5 = R
ch6 = K ch7 = 9 ch8 = r
The output results have already been explained by the comments given in the program. The
characters are represented between single quotes like ‘K’. The computer stores their numerical
value according to the ASCII code.

CONSTANTS
C++ provides many different types of constants. For instance, in the above program we have seen
a character such as ‘A’ is a constant value equal to 65. Similarly ‘K’ is another constant value. These
are character constants. A sequence of characters enclosed between double quotes such as “
Morning” is a string constant.
Similarly, “\n” or “\t’ are constants which have special meaning for compiler and are known
as escape sequences. These are discussed in Chapter 2.
Sequences of digits are integral constants. These may be decimal (base 10), octal (base 8) or
hexadecimal (base 16). A decimal sequence of digits should not start with 0, because, in that case
it would be taken as octal number by computer. Octal numbers are preceded by zero (0). The
digits of the three systems have already been discussed in Chapter 1. If you see page 20 of Chapter
1, we have,
456 is a decimal number. Its equivalents in octal and hexadecimal are given below.
0710 is octal number. In this, 0 on extreme left indicates that it is octal number.
0x1c8 is hexadecimal number. 0x indicates that it is hexadecimal number.
Sequences of digits of base 10 are called decimal constants. Similarly, a digital sequence of
base 8 is called octal constant and a sequence of hexadecimal digits is hexadecimal constant.

THE CONST AND VOLATILE

The const and volatile are attribute modifiers. A quantity, whether it is int, float, double, char
or user defined type (a class object), if it is declared const its value cannot be changed in the
program. One may declare a constant quantity as below.
const int Diameter = 10 ; // the statement fixes the Diameter to 10 units.
const float PI = 3.14159; // const float value which cannot be modified later.
const char kh [ ] = “ Good Morning”; // a constant string of characters.
const wchar_t wch [ ] = L “Trip”; // a constant string of wide characters.
Note that the values in wide characters are preceded by L.
The volatile may be used in definition of a variable which may be modified from outside
the program, It is decleared as volatile int x;

STORAGE CLASS SPECIFIERS

The following five storage class specifiers are supported by C++. These are auto, register, extern,
static and mutable. These are disenssed below.
❖ 64 ❖ Programming with C++

auto : The local variables belong to auto storage class. It is optional to use the specifier auto.
Thus the declaration int n; is same as auto int n ;
The scope of these variables is the block of program (statements enclosed in a pair of curly
brackets {}) in which they are created. During running of program, after the block is over the
variables are removed from the memory and are not available.
static: A static variable may be a local or a global variable. Global variable may be accessed
from any part of the program. Also see the section on scope of variables.
register : A CPU has a number of registers ( 16 is a typical number) for various purposes.
A programmer may use the specifier register in the declaration of a variable if the same is desired
to be loaded on a register so that it is readily (in less time) available to the processor. If the variable
is frequently used in the program, the time of execution of program may decrease by this
specification. However, it is the compiler which will decide if the recommendation can be
honoured.
extern : The specifier extern gives the variable an external linkage and it may be accessed from
another linked file.
mutable : An object of a class may be declared constant. However, if a data member of class
is declared mutable, it may be modified in a constant object.
The following program illustrates extern, static, auto, register and bool data type.
PROGRAM 3.4 – Illustrates different storage class data.

#include <iostream>
using namespace std;
extern int n = 10; // global variable
int y = 5; // global variable
int main()
{ static int D = 5;
int x =7;
auto int z = 25 ;
cout <<“n*n = “ <<n*n << “, y*y = “<< :: y *::y<<endl;
register int m = 5; // m to be loaded on a register
cout << “Product m *D = “<< m*D <<endl;
bool B = (x == y); // B is a bool type variable
bool C = (y*y == 25); // C is also bool type variable
cout << “B = “ << B<< “, C = “ << C<<endl;
return 0;
}

The expected output is given below.

n*n = 100, y*y = 25
Product m *D = 25
B = 0, C = 1

The following program illustrates the static specifier.

 Fundamental Data Types in C++ ❖ 65❖
PROGRAM 3.5 – Illustrates static storage class specifier.

#include <iostream>
using namespace std;
static int n = 10;
int main()
{{ int n = 20;
int y = 8;
cout <<“ y = ”<<y << “, n = ” << n <<“ and ::n = ”<< ::n<<endl;
}
cout << “Outside the inner curly brackets n = ” << n <<endl;
cout <<“And ::n = ”<<::n<< endl;
return 0 ; }

The expected output is given below.

y = 8, n = 20 and ::n = 10
Outside the inner curly brackets n = 10
And ::n = 10

The enum type may be used to define a number of integral constants as discussed on page
56. The following program gives an illustration.

PROGRAM 3.6 – Illustrates another example of enum data type.

#include <iostream>
using namespace std;
int main()
{ int n, m;
double p, q;
enum Day { Mon=1, Tues, Wednes, Thurs, Fri, Sat, Sun};
enum data { A = 10, B=5, C = 6}; // defining integral constants
cout<< “Thurs = “<<Thurs<<endl;
n = Wednes + Mon; // Addition
p = Mon - Fri; // Subtraction
q = Sun/Wednes ; //integer division
m = A*B*C; // using enum data
cout<<“n = “ << n <<“,\t p = “ <<p <<endl;
cout<< “q = “<<q <<endl;
cout<<“Saturday= “ <<Sat<<“\n”;
cout<< “m = “ << m <<endl;
return 0;
}
❖ 66 ❖ Programming with C++

The expected output is given below, enum type has been used to define several integral values.
Thurs = 4
n = 4, p = -4
q=2
Saturday= 6
m = 300

3.6 SCOPE OF VARIABLES

The scope of a variable depends on the storage class to which it belongs, i.e. automatic storage class
or static storage class. The local variables belong to automatic storage class. The keywords auto
(generally not used) and register (if the variable is desired to be loaded on register) may be used
before variable names. The scope of both these is limited to the nearest program block enclosed
by braces { } in which they are defined. Figure 3.1 illustrates declarations of local variables which
belong to automatic storage class and their scope along with the scope of static storage class variables.

int x = 20; //global variable

static double z = 5.6;

int main ()

{ int x = 6;

static int y = 8; // x = 6

{ int x = 8; //x =8 //::x = 20

{ int x = 10;
//x =10 //z =5.6
/* (Values 8 & 6 not visible
} if x is redefined)*/
// y = 8
//x =8
}
return 0 ; // x = 6
}
Fig. 3.1: Scope of variables

In the static storage class there are two categories, i.e. external or global variables and local static
variables. The global variables are defined outside any class or a function declaration. Thus global
variables are defined outside main (). Their scope is from the point of declaration to the end of
program. Inside the main () they may be accessed with scope resolution operator (::). The static
variables are declared with keyword static inside a function or out side a function. The scopes of
 Fundamental Data Types in C++ ❖ 67❖
local and global variables are illustrated in Fig. 3.1. The static data members for classes are discussed
Chapter 12. The data declared const cannot be modified in the program. The declaration may be
inside a function or outside a function. The following program illustrates the different categories.

PROGRAM 3.7 – The program illustrates the scope of variables.

#include <iostream>
int x = 8; // global variable
double y = 2.2; // global variable
using namespace std;
int main()
{ int x , y;
const double PI = 3.14159;

const int z = 6; // Value can not be changed in the program

x = 11; // variable in main()
y = 9 ; // variable in main()
cout<< “In main \t\tx = “<<x <<“, y = “<<y<<“\t PI = “<<PI <<“\tz = “
<<z<<endl;
{int x, y;
x = 5;
y = 4;
cout<<“In outer braces x = “<<x<<“, y = “<<y <<“\t PI = “
<<PI<< “\tz = “<< z << endl;
{
{ int x, y;
x = 20;
y = 30;
cout <<“In inner braces x = “<<x<<“, y = “<< y << “\t PI = “<<PI
<< “\tz = “<< z <<endl;
}
cout <<“In the outer braces x = “ <<x<< “, y = “ << y <<“\t PI
= “<<PI << “\tz = “<< z <<endl;
}}
cout<< “Outside the outer braces x = “<<x <<“, y = “<<y<<endl;
cout << “In global scope x = “ << ::x << “ y = “<<::y <<endl;
// Getting global value of x
return 0;}
The expected output is given below.
In main x = 11, y = 9 PI = 3.14159 z = 6
In outer braces x = 5, y = 4 PI = 3.14159 z = 6
In inner braces x = 20, y = 30 PI = 3.14159 z = 6
In the outer braces x = 5, y = 4 PI = 3.14159 z = 6
Outside the outer braces x = 11, y = 9
In global scope x = 8 y = 2.2
❖ 68 ❖ Programming with C++

In the above output you see that values x=8 and y = 2.2 are global. In the mid portion of
program and in the innermost braces ‘{ }’, x and y have been declared again as x = 20 and y =
30. This is shown in the third line of output. In outer curly brackets the variables x and y have
values 5 and 4 respectively (2nd and 4th line of output). But the scope of values x = 20 and y =
30 is only up to the closing of the innermost curly bracket. After the closure of inner and outer
braces the values of x and y are again 11 and 9 respectively. That is illustrates by the last but one
line of output in which the values of x and y are same as assigned in the beginning of program,
i.e. x = 11 and y = 9, while in between the values are different. The variables PI and z do not
change their values throughout the program because these are declared const. The global values
are again same, i.e. 8 and 2.2. Note that global values have been declared above the main (). For
accessing global variables the scope resolution operator (::) has to be used. The global values may
be obtained any where in the program by using scope resolution operator (::) before the variable
name. However, for local variables the scope are limited to nearest curly braces { } in which they
are defined. The variables defined in a function have scope only up to the function.

3.7 TYPE CASTING

The change in type of a variable whether from one fundamental type to another fundamental
type, or among user defined types, i.e. from an object of one class to object of another class or
change of a pointer from one class to another class, etc., are called type casting. These may be
carried out as below.
(type_now_desired) expression; // C style of type casting
or
type_ now_desired (expression); // C++ style of type casting
// some take this also as c-type casting

Thus if int A is now desired to be changed to double it may be coded as,

(double) A; // C style
or
double (A); // C++ style
The code of following program gives a simple example as to why type casting is required.

PROGRAM 3.8 – Illustrates the need for type casting.

#include<iostream>
using namespace std;
int main()
{int A = 3,B= 4, C =7;
cout<< “ A/B = “ <<A/B <<endl;
cout <<“ C/B = “<< C/B <<endl;
cout <<“ A/(double)B = “ << A/(double)B<<endl;
cout <<“ double (C)/B = “ << double (C)/B<<endl;
return 0;}
 Fundamental Data Types in C++ ❖ 69❖
The expected output is below.
A/B = 0
C/B = 1
A/(double)B = 0.75
double (C)/B = 1.75

Division of int with another int results in int number and because A/B = 3/4 is a fraction,
therefore, the output is 0. This is illustrated by the first line of output. Similarly the second
division 7/4 results in 1 because the fractional part (.75) is neglected in integer division. But
situation can be saved if we convert either numerator or denominator into a double number as
done in the next two lines of output. In one case the denominator is changed to double and in
second case numerator is changed to double. Now the output also contains the fractional part.
More type conversions are illustrated in the program given below.
PROGRAM 3.9 – Program is another illustration of type casting.

#include <iostream>
using namespace std;
int main()
{
int n ;
double k = 3.675 , m1, m2 , p1,p2 ;
char ch = ‘B’;
n = int(ch);
cout << “ch = ”<<ch << “\t\t k = ”<<k<<endl;
k = int(k);

cout << “n = ” <<n << “\t \t k = ”<<k<<endl;

p1 = (int)7.5/3.0 ;
p2 = (int) 7.5/3;
m1 = int (4.5/2);
m2 = int (4.5/2.2);

cout<< “ p1 = ” <<p1 <<“,\t p2 = ”<<p2 <<endl;

cout << “ m1 = ”<<m1<< “,\t m2 = ”<<m2 <<endl;
return 0;
}

The expected output is given below.

ch = B k = 3.675
n = 66 k=3
p1 = 2.33333, p2 = 2
m1 = 2, m2 = 2
❖ 70 ❖ Programming with C++

The output shows that double k has been changed to an integer. The numerical value of
character ch has been assigned to n. Numerical value of ‘B’ is 66. Note that value of p1 has not
been changed to int while value of p2 has been changed. The difference is that value of p1
contains float type operand so it is taken float. Both m1 and m2 values have been changed to
int because the type casting applies to the result of division because of brackets. Therefore, for
conversion of type of the result of a calculation it is better to enclose the calculation part in
brackets.

NEW TYPE CASTING OPERATORS IN C++

In C++ the following four new casting operators are defined.
static_cast < > ( )
dynamic_cast < > ( )
reinterpret_cast < > ( )
const_cast < > ( )

Some of these operators are meant to be used for specific applications. For instance
dynamic_cast<>() is applied for runtime type information in inheritances with polymorphic base
class. Since we have still not learnt about classes, pointers, inheritance and virtual functions, so
the details of all of these cannot be discussed here. These are discussed in Chapter 15.
The application of static_cast <>() is however discussed below as it can be applied to
fundamental types as well to user defined types. Unlike C-type casting in static_cast the
constantness of the object is preserved. As applied to fundamental data types the static_cast <>()
is coded as below.
static_cast <desired_type> (object);
Here we limit the application to the fundamental types like int, char and double. The
following program illustrates its application.

PROGRAM 3.10 – This program illustrates conversion by static_cast <>()

#include <iostream>
using namespace std;

int main ()
{
double n = 69.54, p =5.5, a ,m ;
double s = 4.4563 , q = 2.45, b = 3.23 ;
s = static_cast<int>(s) + static_cast<int>(p);
//s is assigned integral portions of s and p
cout<<“s = “<<s<<endl;

p = static_cast<int>(q);
m=65;
char ch = static_cast<char>(n) ;
// double n is changed to a character
cout<< “ch = “ << ch<< endl;
 Fundamental Data Types in C++ ❖ 71❖
a = static_cast<int>(b)+ s;
// s has already been changed to integer above
cout<< “a = ”<<a <<“, \tp = ” << p <<endl;
cout <<“char m = “<<static_cast<char>(m)<<endl;
cout<< “q = ” << q << “, \t b = ” <<b<<endl;
cout<< “m = ” << m<<endl;
return 0;
}

The expected output is given below.

s=9
ch = E
a = 12, p=2
char m = A
q = 2.45, b = 3.23
m = 65

The first line of output, the s has been assigned the sum of integral parts of s and p. Hence,
s = 9. The double n is cast into a character value. So n is first converted to int, i.e. 69 and then
to character ‘E’ and the ‘E’ has been assigned to ch. Similarly, int m which has value 65 is
converted into a character ‘A’. The output for a , p and m can be similarly explained. The last
two lines of output show that when cast converted values are assigned, the variable themselves
do not change.

3.8 THE typedef

The typedef allows us to create alternative name for a type which may be a fundamental type or
user defined type. For instance, with the following declaration we may simply write D in place
of double.
typedef double D ; // The typedef of double is D
The application is illustrated in the following program.

PROGRAM 3.11 – Illustrates application of typedef.

#include<iostream>
#include<cmath>
using namespace std;

typedef double D ;
void main()
{
D x, y, Area ; // Here D stands for double
cout<< “Enter two point numbers : ”; cin >> x>>y;
❖ 72 ❖ Programming with C++
Area = x*y;
cout<< “Area = “ << Area<<endl;
}

The expected output is given below.

Enter two point numbers : 8.5 10
Area = 85

3.9 THE typeid ( ) operator

The operator typeid () identifies the type of variable. Its real use is with dynamic objects created
in inheritance with virtual functions and base class pointer wherein the information about type
of object ( to which class it belongs) is not available from the pointer (see Chapter 15). For
fundamental types the operator is not of much use because these do not change type. Here we
only show the use of the operator. It returns reference to type of its argument. It is coded as,
typeid(object);
For determining the name of type it is coded as below.
typeid (object).name();
The application is illustrated in the following program.

PROGRAM 3.12 – Illustrates application of operator typeid ()

#include <iostream>
using namespace std;
void main ()
{
int x, y;
double z;
if(typeid(x) == typeid(y))
cout << “x and y are of same type.”<<endl;
else
cout<< “x and y are of different type. “ <<endl;
cout << “The type of z is “<< typeid(z).name()<<endl;
cout<< “The type of y is “<< typeid(y).name() << endl;
}

The expected output of the program is given below.

x and y are of same type.
The type of z is double
The type of y is int
 Fundamental Data Types in C++ ❖ 73❖

3.10 ARITHMETIC OPERATIONS ON VARIABLES

The operators including arithmetic operators are dealt in detail in the next chapter. However,
here we introduce the arithmetic operators. The five arithmetic operators are listed in the Table
3.4 below.
Table 3.4 – Arithmetic operators
S. No. Description of operation Operator symbol
1. Addition +
2. Subtraction –
3. Multiplication *
4. Division /
5. Modulus ( returns remainder on division) %

The following program illustrates some of the arithmetic operations.

PROGRAM 3.13 – Illustrates arithmetic operations

#include <iostream>
using namespace std;
int main()
{ int a = 4,b = 6, E;
double c,d,A,B,C,D;
c = 2.6;
d = 1.3;
A = c + d; // addition
B = c - d; // subtraction
C = a/d; // division
D = d * d; //multiplication
E =b % a; // modulus operator
cout<< “A =”<<A<<“, B = ”<<B<<“, C= ”<<C<<“, D = ”<<D<<“, E = ”<<E<<“\n”;
return 0;
}
The expected output is given below.
A =3.9, B = 1.3, C= 3.07692, D = 1.69, E = 2

3.11 FUNCTION swap()

Function swap( ) is a C++ Standard Library function. It exchanges the values of two variables.
The variables should be of same type. See the following program for illustration.
❖ 74 ❖ Programming with C++

PROGRAM 3.14 – Illustrates application of swap( ) function.

#include <iostream>
using namespace std;
int main()
{//swap function interchanges the values of same type variables
int x ,y;
cout<<“Write two integers x = ”;
cin>>x; cout<<“ and y = “; cin>>y;
swap(x,y);
cout <<“After swapping: x = ”<<x <<“ and y = ”<< y<<endl;
return 0;
}

In this program the user is asked to enter two integers for x and y. The program interchanges
their values. The output of the program is given below.

Write two integers x = 35

and y = 45
After swapping: x = 45 and y = 35

It should be noted that variables used as arguments of function swap () must be of same type.
The function swap() belongs to C++ Standard Library.

EXERCISES
1. What are the fundamental types? Why are these called fundamental types?
2. How do you declare a variable?
3. Which of the following identifiers are incorrect?
(i) 5chairs (ii) Money
(iii) template (iv) typeid
(v) bitor
4. What are keywords in C++?
5. What do you understand by typeid () and typedef?

6. Is it necessary to declare all the variables right in the beginning in a C++ program?
7. What is the difference between the number 5 and the character ‘5’?
8. Can the keywords be used as variable name?
9. Which of the following statements are wrong? Also correct the wrong statements.
(a) Int x = 7; (b) int y = 5.0 ;
(c) double = 4.76548; (d) Float = 5.463;
 Fundamental Data Types in C++ ❖ 75❖
10. Are the following initializations correct? If not, give the correct version.
(a) int x = 6 ; (b) short int y = 8 ;
(c) unsigned long = – 4500; (d) enum (Monday, Tuesday, Wednesday) ;
11. Differentiate between double and long double numbers.
12. What is wrong with following statements? Also give the correct version.
(a) int x = 6.75; (b) int shot = 8;
(c) char m = A; (d) double n = int (B);
13. What information do we get by using the function sizeof()?
14. Make a program to read three integer numbers and to calculate and display their squares and
cubes.
15. Make a program to read, calculate and display the cubes of two small numbers which are
entered by user.
16. Write a program that prints ASCII code for upper case alphabets A to C.
17. Write a program that reads length in inches and converts it into centimetres.
18. Write a program that converts Fahrenheit into centigrade.
19. What does swap() function do? Can it swap values of variables of different types?
20. Write a program to convert character variables into integers and integer variables into character
variables.
21. What do you understand by scope of a variable?
22. Explain the difference between the following.
(i) variable with global scope.
(ii) variable with scope in main().
(iii) variable with local scope.
23. Give examples of code of the above three types of variables.
24. Explain the scopes of global and static variables.
25. What is function scope?
26. What is the difference between a static variable and a constant.
27. Make a small program that asks the user to enter three numbers and displays their cubes.
28. Write the expected output of the following program.

PROGRAM 3.15 – Another example of type-casting.

#include <iostream>
using namespace std;
int main ()
{
int n = 80.54, p =5, a ,m ;
double s = 4.4563 , q = 2.45, b = 3.23 ;
s = int(s) + int(p);
cout<<“ s = “<<s<<endl;
❖ 76 ❖ Programming with C++
p = int(q);
m=65;
char ch = char(n) ;
cout<< “ ch = “ << ch<< endl;
a = int(b)+ s;
cout<< “ a = “<<a <<“, \tp = “ << p <<endl;
cout <<“ char m = “<<char (m)<<endl;
cout<< “ q = “ << q << “, \t b = “ <<b<<endl;
cout<< “ m = “ << m<<endl;
return 0;
}

29. Make a program to illustrate the values of character ‘+’ and size of “A”.

PROGRAM 3.16 – Illustrates the values of characters ‘+’ and size of “A”.

#include <iostream>
using namespace std;
int main()
{ char River [6] = “Ganga”;
char ch[] = “A”;

char Plus = ‘+’;

cout<< “Character + = “ <<int(Plus)<<endl;
cout << sizeof (ch)<<endl;
cout<<River<<“, “ << ch <<endl;
return 0;
}

The expected output is given below.

Character + = 43
2
Ganga, A
❍❍❍
CHAPTER

4.1 INTRODUCTION

Operators are instructions, represented by a single symbol or combination of symbols such as

+, –, /, +=, &, etc., which are used in program statements to carry out some action on the data.
The actions may be assignment, arithmetic operations, comparison of values of two variables,
determination of address of a variable, pointing to memory location or linking an object to its
class member, etc. C++ has a rich inventory of operators, however, in this chapter we shall be
dealing with the following.
(i) Assignment operators
(ii) Arithmetic operators
(iii) Composite assignment operators
(iv) Increment & decrement operators
(v) Relational or comparison operators
(vi) Logical operators
(vii) Bitwise logical operators
Most of these operators are common with those of C. The symbols of these operators are
shown in Table 4.1. Some of these we have already used in previous chapters, however, we have
not examined their behaviour in detail.
C++ supports unary, binary and ternary operators. A unary operator takes one operand for
its operation, a binary operator takes two operands and a ternary operator takes three operands.
There is only one ternary operator in C++, i.e. selection operator (?:) which is discussed in the
next Chapter. Examples of operators which are both unary and binary are + and –. These
operators may be used to change the sign of a value in which case these are unary operators. These
may also be used to add and subtract two quantities in which case they operate on two quantities
and are binary. The arity of an operator is the number of operands it takes for its operation.

OPERATOR PRECEDENCE AND ASSOCIATIVITY

Consider the following simple calculation.
8 - 2*3
❖ 78 ❖ Programming with C++

We know that its result is 2. But how is that result calculated? We first multiply 2 and 3
and then subtract the product (6) from 8. That means that multiplication has higher priority
or precedence than the minus, otherwise if we first subtract 2 from 8 and then multiply the result
would be 18. Each operator has a precedence level which decides its turn for operation when
there are several operators in a statement. The operator with highest precedence is performed first
followed by the one with the next lower precedence. The precedence levels of the operators are
set in the complier and computer follows these rules during calculations.
Now if a number of operators having the same precedence level are there in a statement then
how to decide as to which should be performed first. For example consider the following
calculation.
20 % 3 * 5
The operators % and * have same level of precedence and both are binary operators. In the
previous chapter we studied the modulus operator (%) which gives the remainder on division. Thus
20 % 3 = 2. In the above expression, if operation 20%3 is performed first the result is 10, otherwise
if 3*5 is performed first the result is 5 because in that case 20 %15 = 5. In such cases the
associativity of the operators comes to rescue. Since the operators are binary if the operations are
performed left to right the result is 10 and if it is performed right to left the result is 5. The
associativity of both % and * operators is left to right and thus the correct result is 10.
The operators in C++ are implicitly defined for variables of fundamental types. For user
defined types such as class objects, C++ allows to redefine their functionality so that they can
be used for class objects as well. The examples are (i) manipulations of vectors, (ii) manipulation
of complex numbers, etc. For such cases the operators are overloaded (functionality is redefined)
Even if the functionality of an operator is redefined (see Chapter 13 ) the arity, the precedence
and the associativity of the operator do not change.

4.2 ASSIGNMENT OPERATOR

The basic assignment operator is ( = ) which is often called equal to. On the left of this operator
we write the name of variable to which a value is to be assigned or l-value and on right side we
write the value to be assigned to it or r-value. The l-value is the memory space in which the r-
value is stored. Consider the following assignments.
int x = 5; // Declaration and initialization
int y = 10;
y = x ; // assignment
In the last statement, will the value of x be assigned to y or the value of y be assigned to x?
The associativity of = operator is right to left, so value of x will be assigned to y. In this operation
y becomes 5.
In the previous chapters we have already studied the two methods of declaration and
initialization, which are illustrated below.
type identifier = value ;
type identifier ( value) ;
 Operators ❖ 79❖

The example are given below.

int x = 5, n = 3;
We may write such statements as given below.
int x , Age; // declaration
float y ; // declaration
y = 5.764; // assignment
x = 5; // assignment
Age = 22; // assignment
The second method is illustrated below. .
int Age(22) ; // Age is declared and initialized to 22.
Similarly a floating point variable may be declared and assigned as below.
float PI(3.14159);
double PI(3.1415926535);
Let Kh be name of a variable of type character. We want to assign the character ‘B’ to it.
We may write the code as below.
char Kh;
Kh = ‘B’;
Or we may combine the above two statements as below.
char Kh = ‘B’;
or as,
char Kh(‘B’);

Similarly a string of characters may be assigned. In the following the <string> header file is
included in the program in order to declare variables of type string.
#include<string>
void main()
{string myname ; // myname is a variable of type string
myname = “Monika”; /*”Monika” is the value assigned to myname */
string Name (“Sujata”); // Name is assigned value “Sujata”
string name = “Mamta”; }
However, if the header file <string> is not included, a string variable may be declared as C-
string. For such a case the assignment is illustrated below.
char myname [] = “Monika”;
The constant 0 (zero) is of type int, however, it may be assigned to variables of all types.
Thus we can write as below.
❖ 80 ❖ Programming with C++
int x = 0; // It assigns 0 to int x.
float y = 0; // It assigns 0.0 to float y
double d = 0; //It assigns 0.0 to double d
char ch = 0; //It assigns Null character ’\0‘ to ch,

Table 4.1 – Operators common to C and C++

Operator Description of Operator Description of

symbol action symbol action

Arithmetic operators , Evaluate

+ Addition Increment/
decrement
– Subtraction ++ Increment by 1
* Multiplication –– Decrement by 1
/ Division Logical operators Also see chapter5
% Modulus && AND
Assignment operators || OR
= Assignment ! NOT
+= Add and assign Bitwise operators
–= Subtract and assign & AND
*= Multiply and assign | OR
/= Divide and assign ^ XOR
%= Assign remainder >> Shift right
Relational operators Also see chapter 5 << Shift left
== Equal to ~ complement
> Greater than &= Bitwise AND assign
< Less than |= Bitwise OR assign
!= Not equal ^= Bitwise XOR assign
>= Greater than or equal <<= Bitwise shift left
and assign
<= Less than or equal >>= Bitwise shift Right
and assign
?: Conditional operator

Programs 4.1 and 4.2, given below, illustrate the assignment of different types of variables.

PROGRAM 4.1 – Illustrates different methods of assignments.

#include <iostream>
using namespace std;
 Operators ❖ 81❖
void main()
{int P(20); // This is equivalent to int P = 20;
int M = 30;
cout<<“P = “<<P<<“,t M = “<<M<<endl;
double C (4.56), D ; // double C(4.56) is equivalent to
// double C = 4.56
D = 6.54;
cout<< “C = “<< C<<“,t D = “ <<D<<endl;

char Kh(‘B’); // For assigning single char the single

// quotes ‘ ’ are required
char ch = ‘A’;
cout << “Kh = “<< Kh << “,t ch = “<< ch <<endl;
}

The expected output is as below

P = 20, M = 30
C = 4.56, D = 6.54
Kh = B, ch = A

The following program makes use of <string> header file which provides similar assignment
operations as we do for int or char variables.

PROGRAM 4.2 – For a string of characters we may use class string (see Chapter 17).

#include <iostream>
using namespace std;
#include <string>
void main()
{ // In following Name1, Name2 and Name3 are names of strings.
string Name1(“Mona Lisa”); // string of characters have to be
// put in double quotes “ “
string Name2 = “Gyatri”;
string Name4 = “Malhotra”;

string Name3;
Name3 = “Raman”; //Assignment = can be used for objects
//of class string.
cout << “Name1 = “<< Name1 <<endl;
cout << “Name2 = “ << Name2 <<endl;
cout<< “Name3 = “<< Name3 <<endl;
cout << “Name2 + Name4 = “ << Name2 + Name4<<endl;
}
❖ 82 ❖ Programming with C++

The expected output of the program is given below. Note that we have used header file <string>
in the program. The objects of this class may be added as illustrated in last line.
Name1 = Mona Lisa
Name2 = Gyatri
Name3 = Raman
Name2 + Name4 = Gyatri Malhotra

4.3 ARITHMETIC OPERATORS

We have already discussed the arithmetic operators in Chapter 3. The different arithmetic
operators are +, –, *, / and % which represent the addition, subtraction, multiplication, division
and modulus. The application of these operators has also been illustrated in Program 3.13 in the
previous chapter. The operator % is applicable only to integers and is generally used to test the
divisibility of a larger integer by a smaller integer. If on division the remainder is zero then larger
number is divisible by the smaller number. The following program illustrates the test of
divisibility.

PROGRAM 4.3 – Illustrates the test of divisibility of integers.

#include<iostream>
using namespace std;

void main()
{ int A,B, m;

cout<< “Enter two integers :”; cin >> A >>B;

cout << “you have entered A = ”<<A <<“ and B = ” << B<<endl;

if (A %B ) //the condition is if A%B is more than zero

cout<<“A is not divisible by B” <<endl;
else
cout <<“A is divisible by B “<<endl;
}

The expected output is as under. Two numbers 64 and 2 are entered. 64 is divisible by 2.
Enter two integers :64 2
you have entered A = 64 and B = 2
A is divisible by B

The following program calculates the roots of a quadratic equation ax2 + bx + c = 0. Because
this involves the evaluation of square root so the header file <cmath> is included in the program.
 Operators ❖ 83❖
PROGRAM 4.4 – The program calculates roots of a quadratic equation.

#include <iostream>
#include <cmath> //<cmath> header file is required for
//calculating square root
using namespace std;
int main()
{ // quadratic equation is: a x2 + bx + c = 0
double a,b,c, P, S;
cout << “Enter the values of a,b and c :”;
cin>>a>>b>>c;

P = b*b-4*a*c ;

if( P < 0) // If P is –ve, the roots are imaginary.

{ P = -P;
S = sqrt(P)/(2*a);
cout<<“Root R1 = “<< –b/(2*a) <<“+” <<“i”<<S << “ and Root R2 =
“<< –b/(2*a)<<“–” <<“i”<<S<<endl;
}
else
{ S = sqrt(P) ; // If P is +ve the roots are real.

cout <<“Root R1= “<< (–b+ S)/(2*a) <<“, and Root R2 = “<< (–b –
S)/(2*a) <<endl;
}
return 0;
}

The following two trials of the above program are carried out with different values of a, b and
c. In the first case imaginary roots are obtained and in second case real roots are obtained.
Enter the values of a,b and c :4 10 10.25
Root R1 = –1.25+i1 and Root R2 = –1.25–i1
The following real roots are obtained when we enter a = 4, b = 10 and c = 4.
Enter the values of a,b and c :4 10 4
Root R1= –0.5, and Root R2 = –2

PRECEDENCE IN ARITHMETIC OPERATORS

All the arithmetic operators do not have the same precedence. Computer first evaluates those with
highest precedence then it evaluates those with next precedence and so on. Table 4.2 gives
precedence of arithmetic operators.
❖ 84 ❖ Programming with C++
Table 4.2 – Arithmetic operators

Arithmetic operation C++ operator Order of evaluation

Parentheses () First. If there are several nested pairs,
the inner most pair is evaluated first.
If in parallel, from left to right.
Multiplication, division, * / % Equal. If there are several than left to
modulus right
Addition, Subtraction + – Equal, If there are several from left to
right

Below is an illustration of how an expression for A is evaluated by computer.

A = 5*((3 + 4) +6) – 2 *(3+4);
= 5*(7 + 6) – 2*(3+4)
= 5*13 – 2*(3+4)
= 5*13 – 2*7
= 65 – 2*7
= 65 – 14
= 51
In some cases when the expressions are not written carefully the precedence may give an
unexpected result. This is demonstrated by following program.

PROGRAM 4.5 – Illustrates the effect of precedence on the value calculated.

#include <iostream>
using namespace std;

int main()
{
int a = 15, n = 3,s = 8, p = 6,r = 3,A ,B , P1,P2,P3 ;
double C,D;
A = n*a%p ; // here first n and a are multiplied before
//the operation of %.
B = n*(a%p); // here () is evaluated first and the result
//is then multiplied to n.
P1 = n% a*p ;
P2 = (n%a)*p ;

P3 = n%(a*p);

C =(n +s +p +a)/r; //bracket is evaluated first

 Operators ❖ 85❖
D = n + s+p+a/r; //Here a/r is evaluated first. It is
// then added to n+s+p.
cout<< “A = “ << A <<“\t B = “ << B <<endl;
cout<< “P1 =” <<P1<<“\t P2 = “ << P2 <<“\tP3 = “<< P3<<endl;
cout <<“C = “<< C <<“\t D = “ << D << endl;
return 0;
}

The expected output is given below.

A = 3 B = 9
P1 =18 P2 = 18 P3 = 3
C = 10 D = 22
The above program emphasizes how the precedence can make a difference in the calculation.
Let us take the line,

A = n * a % p ;
Though the precedence level of * and % is same but evaluation is from left to right so the
above expression is equivalent to the following.
A = (n * a) % p ;
Therefore, the expressions A = (n * a) % p; and A = n *( a % p); would not give same
result. Same is true for evaluation of P1, P2 and P3. The expression P1 = n%a*p ; is equivalent
to the following.
P2 = (n%a)*p ;
But P3 = n%(a*p); is different because the priority has been changed. In this the expression
(a*p) is evaluated first.

4.4 COMPOSITE ASSIGNMENT OPERATORS

C++ allows combining the arithmetic operators with assignment operator. The arithmetic
operator is written first followed by assignment operator but not vice versa. Thus we write the
code for add and assign as below,
B += 6;
which means that first add 6 to the present value of B and then assign the result to B. This is
equivalent to the following statement.
B = B + 6;
Similarly the other operators may also be combined. All the composite operators are listed
in Table 4.3. The applications of some of these are given in the Program 4.6.
❖ 86 ❖ Programming with C++

Table 4.3 – Composite assignment operators

Operator C++ expression Equivalent expression Explanation and use

+= B += 5; B = B +5; int B= 4;
B +=5 ; // makes B=9.
–= C –= 6; C = C – 6; int C = 10;
C –= 6 ; // makes C=4
*= D *= 2; D = D*2; int D = 10;
D *=2; // makes D=20
/= E /= 3; E = E/3; int E = 21;
E /=3; //makes E = 7
%= F %= 4; F = F % 4 ; Divides F by 4 and remainder
is assigned to F

In all the composite operators there should not be any space between the symbols. For instance,
if we give space between + and = such as + = it may result in error.

PROGRAM 4.6 – The program illustrates the application of composite operators.

#include <iostream>
using namespace std;
int main ()
{
int m =3, n=4, p=5, s = 6, q = 11;
m += 2; // no space between + and =
n *= 4;
p –= 3;
s /=2;
q %=4;
cout <<“m = “<<m <<“, n = “<<n<<“, p = “<<p<<“, s = “ <<s<<endl;
cout << “q = “ << q << endl;
return 0;
}

The expected output is given below. The out put is self explanatory.
m = 5, n = 16, p = 2, s = 3
q = 3

4.5 INCREMENT AND DECREMENT OPERATORS

For increasing and decreasing the values of integral objects by unity, we may make use of
increment operator ++ and decrement operator – – respectively. These operators may be placed
before or after the variable name. When the operators ++ and – – are placed before the variable
name, these are called pre-increment and pre-decrement operators respectively. When the
operators are placed after the names then these are called post-increment and post-decrement
operators. In case of pre-increment the present value of variable is first increased by unity and
the incremental value is then used in the application. And in case of post-increment, the present
 Operators ❖ 87❖

value of variable is used in the application and then the value of variable is incremented. Same is
the case of operator – – . The pre and post versions of ++ and – – are also elaborated in Table 4.4.
Also take care that there should not be any space between ++ or between – – because this may
result in error.
Table 4.4 – Increment and decrement operators

Pre or post Operator Description

Pre-increment ++ k First increase the value of k by unity. Then evaluate
current statement / function by taking incremented
value.
Post-increment k++ First use the current value of k to valuate current
statements / function, and then increase k by unity
Pre-decrement – –k First decrease the value of k by unity and assign it
to k. Then evaluate the statements / functions.
Post-decrement k– – First use the current value of k to valuate current
statements / functions then decrease k by unity.

The following program illustrates the increment and decrement operators.

PROGRAM 4.7 – Illustrates increment and decrement operators.

#include <iostream>
using namespace std;
int main()

{int a = 6, p = 4,r=3,n =5,A,B,C,K ;

A = 6*++n ;
cout << “A = “<<A <<“\t n = “ <<n <<endl;
K = 5*a– – ;
cout<<“K = “<<K<<“\t a = “ <<a << endl;

B =r++*r++ ;
cout<< “B = “<<B<<“\t r = “<< r << endl;
C = p– – *p– – ;
cout<<“C= “<< C<<“\t p= “ << p << endl;
return 0;
}

The expected output given below is self explanatory. However, it is also explained.
A = 36 n = 6
K = 30 a = 5
B = 9 r = 5
C= 16 p= 2
❖ 88 ❖ Programming with C++

For the first line of output n = 5. The n is first incremented to 6 and then multiplied by 6
to give 36. The second line of output is obtained by multiplication of 5 with the initial value
of a, which is 6. The variable a is then decremented to 5. The third line is obtained by
multiplying r with r, taking the initial value. Then r is incremented twice. The last line of output
is obtained similarly. The program below illustrates more such cases.

PROGRAM 4.8 – Illustrates application of increment and decrement operators.

#include <iostream>
using namespace std;
int main()
{int m=6,n=2,p=4,r=3,s=8,k = 4,a,b,c,d ;
a =++n+ ++m;
b = s++ * s++;
c = p– – *p– – ;
d = (– – k* – – k)*++r ;
cout<<“a = “<<a<<“,\tn = “<<n<<“,\tm = “<<m<<endl;
cout<<“b = “<<b <<“,\ts = “ <<s<<“\n”;
cout<<“c = “<<c<<“,\tp = “<<p<<endl;
cout<<“d = “<<d <<“,\tk = “<<k<<“,\tr = “<<r<<endl;
return 0;
}

The expected output is given below.

a = 10, n = 3, m = 7
b = 64, s = 10
c = 16, p = 2
d = 16, k = 2, r = 4
In the first line of output a = (2 + 1) + (6 + 1) = 10. For the second line of output s is
increased after the multiplication, so b = 8 × 8 = 64. But s has got increment twice, so its value
becomes 10. Similarly in the third line of output the decrement in the value of p happens after
the multiplication, therefore value of c = 16 and p has got decrement twice, so its value has
decreased from 4 to 2. For the fourth line of output the value of k is pre-decreased twice because
the decrement operator has higher precedence than the operator *. Similarly value of r is increased
from 3 to 4 then multiplied with the first factor which is 4. So the result is 16.
As you have seen in the above program, in a complex expression wherein a variable appears
several times, it would be better not to use increment or decrement operators because it becomes
difficult for the programmer to keep track of changes in values. In the following program for
the expression E = ++a*– –a*++a; you would expect for a = 6, the E = 7 × 6 × 7, but it is equal
to 6 × 6 × 7. Because value of a is first increased to 7 and then decreased to 6 before the first
multiplication. It is then increased to 7 and multiplied to 36.
Similar cases can arise in expressions like ++m*– – m. If initial value of m is 4 you would
expect the expression to evaluate to (4 + 1)(5 – 1) = 20 but program gives a value 16. Because
the value of m is first increased then it is followed by decrement and then multiplication. So
the expression evaluates as 4 × 4 = 16. Thus if we have an expression like
 Operators ❖ 89❖

A = ++n * – – n ;
The result is A = n*n ; with the initial value of n, because ++ and – – have higher precedence
than *. But if you have an expression like
A = ++n *– – p* – – n;
It would evaluate as A = (n+1) *(p–1)*(n–1);
The following program involves product of three factors like the example given above, and
it is interesting to note the results. Would you like to explain the results?

PROGRAM 4.9 – Illustrates use of operator ++ and – – in along with other operators.

#include <iostream>
using namespace std;
int main()
{
int a = 6, n= 5,s =8, p=4,r=3,A,B,C,K,M,D,E ;
A = ++n*– – p * – – n ; // Equivalent to 6*3*5=90 and now p = 3
cout << “n = “<<n <<“, A = “ <<A <<endl;
B = ++a*2*– – a; // This is equivalent to 7*2*6
D = – – a * ++a ; // This is equivalent to 6*6
E = ++a*– – a*++a; // Equivalent to 6*6*7 = 252
cout<<“B = “<<B<<“, a = “ <<a << “, D = “<<D <<“, E = “ << E
<<endl;

M = ++s*– – s; // equivalent to 8*8

K =++r +4 – – r ; // Equivalent to 4+4-3=5

cout<< “K = “<<K<<“, r = “<< r <<“, M = “<<M<< endl;
C =p++* 5*p––; // Equivalent to p *5 *p and p =3
// See calculation for A above
cout<<“C= “<< C<<“, p = “ << p << endl;
return 0;
}
The expected output is given below.
n = 5, A = 90
B = 84, a = 7, D = 36, E = 252
K = 5, r = 3, M = 64
C= 45, p = 3

4.6 RELATIONAL OPERATORS

These operators are used to compare two variables and are generally used with conditional
statements. For example, the code if(A==B) means if A and B are equal. Note that in this two
‘equal to’ signs (==) are used. Programmers often make a mistake by using single =. The other
relational operators are explained in Table 4.1. These are explained as well as their applications
are illustrated in Chapter 5.
❖ 90 ❖ Programming with C++

4.7 BOOLEAN OPERATORS

Boolean operators are logical operators. They relate the logical statements. Let A and B be the
two variable representing two logical statements. The logical variable have only two states that
is either it is true or false. If it is true the variable has a value 1 or a positive value. If it is false it
has value zero. The three Boolean operators are OR, AND and NOT. Table 4.5 gives details
about these operators and their symbols.
Table 4.5 – Boolean Operators

Condition Operator Description

(i) A OR B || If either A or B or both are true then A|| B =1
It is false only when both A and B are false. In that
case A||B = 0
(ii) A AND B && If both A and B are true then A && B = 1 else
A&&B =0
(It is false when any one is false)
(iii) NOT A ! True if A evaluates to be false, i.e.
If A is false then !A = 1
and false when A evaluates true, i.e.
If A is true then !A = 0

The evaluation of various conditions in terms of true or false in different situations are given
in Table 4.6(a,b,c) below.

Table 4.6 – (a), (b) and (c)

(a) A OR B (b) A AND B (c) NOT A

A B A||B A B A&&B A !A
false false false false false false true false
true false true true false false false true
false true true false true false
true true true true true true

The program below illustrates the applications of Boolean operators.

PROGRAM 4.10 – Illustrates the application of Boolean operators.

#include <iostream>
using namespace std;
int main()
{ int p, q, r, s,t,x, y, z;
p =1;
 Operators ❖ 91❖

q = 0;
r = 1;

s = p||q;
t = !q;
x = p&&q;
y = (p || q && r||s);
z = (!p || !q && !r || s);

cout << “s = ”<<s << “, t = ” << t <<“, x = ”<<x<< endl;

cout << “y = ” <<y <<“, z = ”<<z<< endl;
return 0; }

The expected output given below should be verified by reader by calculations with help of Table
4.7(a,b,c).
s = 1, t = 1, x = 0
y = 1, z = 1

4.8 BITWISE OPERATORS

As already discussed in Chapter 1, a bit is the basic unit of memory. It can have only two states.
If it is set its value is 1 otherwise it is zero. Bitwise operation means convert the number into
binary and the carry out the operation on each bit individually. For example let us take the
operation complement represented by symbol (~). Let us take a number
short A = 42;
A short number is stored in two byte or 16 bits. Therefore,
A when expressed in binary = 00000000 00101010
Complement of A is ~A = 11111111 11010101
The compliment operator changes the bit with value 0 to 1 and the one with value 1 is
changed to 0. The different bitwise operators are listed in Table 4.7 below. It also illustrates
codes.
Table 4 .7 – Bitwise operators

Operator Description Example of code

& AND A&B ;
| OR A|B;
^ XOR A^B ;
>> Shift right A >>2;//shift right by 2 places
<< Shift left A<<2;
~ Complement ~A;
&= Bitwise AND assign A &= B;

Contd...
❖ 92 ❖ Programming with C++

|= Bitwise OR assign A |= B;
^= Bitwise XOR assign A ^= B;
<<= Bitwise shift left assign A <<= 2;
Shift left by 2 places and assign to A
>>= Bitwise shift right assign A >>= 1;
Shift right by 1 place and assign to A

The operators AND, OR and XOR are explained in truth Table 4.8 below.
Table 4.8 – Bitwise operators AND, OR and XOR

Bit1= A Bit2 = B A&B A|B A^B

0 0 0 0 0
1 0 0 1 1
0 1 0 1 1
1 1 1 1 0

The programs given below explain the application of some of these operators.

PROGRAM 4.11 – Operator complement (~) is used to determine the maximum value of
an integer that can be stored on my PC.

#include<iostream>
using namespace std;

void main()
{ unsigned A = 0;
A = ~ A;
cout << “A = “<<A <<endl;
}

The expected output is given below.

A = 4294967295
The following program illustrates the application of bitwise AND operator (&).

PROGRAM 4.12 – Illustrates bitwise AND operator.

#include<iostream>
using namespace std;

void main()
{ short A =24;
 Operators ❖ 93❖
short B = 8;
int C = A&B;
cout << “C = “<<C <<endl;
}
The expected output is given below.
C = 8
The output is explained as below. A short number is stored in two bytes, therefore,
24 in binary is equal to 00000000 00011000
8 in binary is equal to 00000000 00001000
Bitwise AND is equal to 00000000 00001000
The result 0000000000001000 in binary is equal to 8.
The following program illustrates OR and left shift operators.

PROGRAM 4.13 – Illustrates the operators OR and left shift (<<).

#include<iostream>
using namespace std;
void main()
{ short A =42;
short B = 12;
int C = A|B;
int D = A<<1;

cout << “C = “<<C <<endl;

cout<< “D = “<<D <<endl;
}

The expected output is as under.

C = 46
D = 84
Explanation of output is given below. A short number is stored in two bytes or 16 bits.
Therefore,
42 in binary is equal to 00000000 00101010
12 in binary is equal to 00000000 00001100
Bitwise OR operation results in 00000000 00101110
This is equal to 46 in decimal
For the variable D, the digits of A are shifted left by 1 place and assigned to D. So the original
binary number 00000000 00101010 now becomes 00000000 01010100 which is equal to 84,
i.e. the number gets multiplied by 2. Thus shifting a digit to left means multiplication by two.
❖ 94 ❖ Programming with C++

PROGRAM 4.14 – Illustrates the operators ^ , >>= and <<= .

#include<iostream>
using namespace std;

void main()
{short A =42;

short B = 12;
short C = 24;
short D = A^B; // XOR operator
C <<= 1;
A <<=2; // Shift to left by 2 places and assign
B >>=2 ; // shift right by 2 places and assign
cout<< “A = ”<<A<< “ \tB = “<< B <<endl;
cout << “C = ”<<C <<endl;
cout << “D = ”<< D <<endl;
}

The expected output is given below.

A = 168 B = 3
C = 48
D = 38

The binary equivalent of A = 42 has been shifted two places to left. So the number gets
multiplied by 4. So A = 168. Similarly the right shift by two places in case of B divides it by
4.The binary equivalent of 24 has been shifted to left by one place and assigned to C. Therefore,
C = 24×2 = 48. For A ^ B see the following explanation.
Binary equivalent of 42 is 00000000 00101010
Binary equivalent of 12 00000000 00001100
Bitwise XOR operation gives 00000000 00100110
This is equivalent to 38.
The following Program illustrates more of the compound assignment operators.

PROGRAM 4.15 – Illustrates the application of operators ^=, &= and |=.

#include<iostream>
using namespace std;
void main()
{ short A =20, D , E, F;
short B = 18;
short C = 30;
D = C^=B;
E = A &=B ;
F = C|=A ;
 Operators ❖ 95❖
cout <<“E = ”<<E <<endl;
cout <<“F = ”<< F <<endl;
cout << “D = ” <<D <<endl;
}

The expected output is given below. Try to explain the output.

E = 16
F = 28
D = 12

Chapter 18 deals exclusively with operations on bit sets based on functions in <bitset> header
file.

4.9 PRECEDENCE OF OPERATORS

Precedence is an important aspect of operators. A partial list of operators and their precedence
is given in Table 4.9. The table does not include all the operators because we have not so far dealt
with them. For a more detailed list see Appendix C.
Table 4.9 – A partial precedence table. Those higher in list have higher precedence

Operator Type Associativity

() Parentheses Left to right
[] Array subcript Left to right
++ – – Unary postfix Left to right
++ – – Unary prefix Left to right
+ – Unary plus, unary minus Right to left
* / % Multiplicative Left to right
+– Additive Left to right
<< Bitwise left shift Left to right
>> Bitwise right shift Left to right
<, <=, > , >= Relational operators Left to right
== , != Relational Left to right
& Bitwise AND Left to right
^ Bitwise XOR Left to right
| Bitwise OR Left to right
&& Logical AND Left to right
|| Logical OR Left to right
= , *= , /= , %=, + =, – = Assignment operators Right to left
&=, ^=, |=, <<=, >>= Right to left
Comma Evaluate Left to right
❖ 96 ❖ Programming with C++
EXERCISES
1. State step by step the sequence of operations that a computer would follow to evaluate the
following expressions. (Hint - use precedence of operators)
(a) Z = 5 y + 3y( 10y + 5/2) for y = 2.
(b) Z = 7 y % 2 + 2( 3 + ( y % 3 + 2)) for y = 19
2. State the step by step order in which a computer would evaluate the following expressions?
(Hint - use precedence of operators)
(a) 5 * 5 *( 4 + 6*8) +4;
(b) 7 % 3 *( 5 + 6*7) –5;
(c) 8 * 3 % 5 *(3 * 5 % 2 – 6) –4;
3. What is wrong with the following codes in a program.
(a) n*n( 12 – n)( n + 3); (b) m % 3 (4 – m);
(c) p *p( m + n);
4. Evaluate the following expressions for int n = 10;.
(a) n * n % 3; (b) n*(n % 3);
(c) (n*n) % 5; (d) n/3;
5. Explain the difference between pre and post increments, i.e. ++ n and n++.
6. Evaluate the following for int n = 8 ;
(a) ++n += 3; (b) –n % 3 += 2;
7. Evaluate the following for int n = 4;.
(a) ++n*n++; (b) ++n*n– – ;
(c) 2*++n + 3 * – –n; (d) ++n += 3;
8. Evaluate the following expressions for int n = 20 ;
(a) n /= 5; (b) n += 8;
(c) n %= 7;
9. Write a program to find roots of a quadratic equation.
10. Write a program to numerically prove that sin(A + B ) = sinA .cosB + cos A. sinB.
11. Write a program to prove numerically that (A + B )2 = A2 + B2 + 2 A B.
12. Write a program which reads two integers numbers x and y and carries out the following bitwise
operations on the numbers.
(i) x & y; (ii) x |= y;
(iii) x <<= 2;
13. Write a program to read 5 integers and to display their average value.
14. Write a program to display the average value and standard deviation of 10 integers.
15. Compare the precedence levels of following operators.
+,–,/,*
16. What would be the value of following expression for int m = 10; in the following code?
(i) m – – = ++m;
(ii) m *= ++m;
 Operators ❖ 97❖
17. Make a program to evaluate the following expressions for m = 6, n = 2 and (i) a = 0,
b = 0, c = 0, d = 0, and e = 0. Repeat the same when (ii) a =1, b =1, c = 2, d = 2, and
e = 2.
(i) a += 4 + ++m*n ;
(ii) b *= 3 + – – m*m ;
(iii) c += 2 +m *++m ;
(iv) d *= 2* + m*m– – ;
(v) e – –= 2*++m / m– – ;
Answer –

PROGRAM 4.16 – An exercise in application of increment/decrement operators.

#include<iostream>
#include<cmath>
using namespace std;

int main ()
{
int m = 6,n = 2, a=0,b=0,c=0,d=0,e=0;
a +=4 + ++m*n ;
b *=3+ —m*m ;
c +=2 +m *++m ;
d *=2* + m*m–– ;
e -=2*++m / m–– ;

cout<<“a = “<<a<<“, “<<“b = “ << b<<“, c = “<<c<<“, d

= “<<d <<“, e = “<<e <<endl;
return 0 ;
}

The expected output is as under.

a = 18, b = 0, c = 51, d = 0, e = –217

For the second case the values are

a = 19, b = 39, c = 53, d = 196, e = 0

18. Determine the output of following program.

PROGRAM 4.17

#include <iostream>
using namespace std;
int main ()
❖ 98 ❖ Programming with C++
{
int m=3, n=4, p=5, s=6, q=31;
m += 2;
n *= 4;
p –= 3;
s /= 2;

cout <<“m = ”<<m <<“, n = ”<<n<<“, p = ”<<p<<“, s = ”

<<s<<endl;
q %= 4 + (n /= 4);

cout << “ q = “ << q << endl;

return 0;
}

Verify your answer with the following output obtained by running the program.

m = 5, n = 16, p = 2, s = 3
q = 7
❍❍❍
CHAPTER

5.1 INTRODUCTION

Often we compare two or more items for characteristics such as size, number of items or quality,
etc. For instance, if you have to select a bag out of two bags A and B, you may compare their
sizes, say bag A is bigger than bag B. If you are looking for a bigger size, you may select A and
discard B. In many real life situations our decision for further action depends on the result of
such comparisons. In automatic control systems also the current conditions (values of parameters)
for a process are compared with the set conditions and the differences decide the remedial actions.
In programming for real life problems often the values are compared and the next step is decided
by the outcome of comparison. In C++ like in other programming languages there is a set of
operators which are used for the comparison of characteristics of two objects. These are listed
in the Table 5.1 given below.

Table 5.1 – The operators for different relational (comparison) expressions

Condition Operator Example C++ code with if

(i) equal to == if m is equal to n if (m == n)
(ii) not equal to != if m is not equal to n if (m != n)
(iii) greater than > if A is greater than B if (A > B)
(iv) less than < if C is less than D if (C < D)
(v) less than or equal to <= if M is less than or equal to N if (M<= N)
(vi) greater than or equal >= if m is greater than or equal to n if (m >= n)

When we compare whether two numbers are equal or not we use (==) and not (=) because
(=) is assignment operator. In the code m = n ; the value of n is assigned to m. Thus the operator
(=) does not do any comparison. For comparison whether m is equal to n or not we must use
==. In the second line of the Table 5.1 we have a condition if m is not equal to n. The operator
(!=) is opposite of the operator (==). Conditional comparisons may occur in a variety of
problems. Generally comparison operators are used along with conditional expressions as
illustrated in the last column of the table.
❖ 100 ❖ Programming with C++

5.2 CONDITIONAL EXPRESSIONS

In C++ three conditional expressions are provided. These are (i) if, (ii) if .... else and
(iii) switch .The if condition gives the option of single choice, i.e. if the conditional expression
is true the statement following the condition is carried out. If the condition is false the following
statement is discarded and program goes to next statement. In if .... else there are two choices.
One choice is with if and the second choice is with else. If the conditional expression is true the
statement following if is carried out otherwise the statement following else is carried out. In switch
condition more than two choices are generally provided. There is no limit to the number of
choices. Each switch condition may have its own choice condition. Similar results may also be
obtained by using a chain of if .... else statements. All these are explained below.

5.3 THE if EXPRESSION

We often say like this “if you get marks equal to 50% or more than 50% you can get admission
in the course”. Another similar statement would be “If you know programming with C++ you
are eligible for this job.” A close examination of above expressions shows that the first part of
the sentence is the condition prefixed with if , i.e. if this happens or if it is true then the second
statement will follow. In case, the condition is not true the following statement is discarded and
program proceeds to next statement. In a program it is written as given below.
if (conditional expression )
statement;
The first line of code is the condition. If the expression in the parentheses is true, the
statement following the expression would be carried out, otherwise the statement would be
ignored and the program would proceed further.

Test
True
if (expression)

False Execute next

statement

Skip next statement

To other statements

Fig. 5.1: Execution of if condition

Figure 5.1 illustrates the execution of if condition. You would also note that the conditional
expression is put in brackets and there is no semicolon at the end of line, while the following statement
ends with semicolon. The condition mentioned above has been used in the following program
in which the user is asked to enter the marks obtained by him / her. If marks are more than or
 Selection Statements ❖ 101❖
equal to 50%, the next line of the program will be executed, i.e. it will be displayed on the
monitor. If the expression is not true nothing is displayed on monitor.

PROGRAM 5 .1 – Illustrates if condition

#include<iostream>
using namespace std;
int main()
{ int Marks;
cout<<“Write your marks: ”;
cin>>Marks;
if (Marks >= 50) // conditional expression
cout << “You are eligible for admission.”<< endl;
return 0;
}

As you click for execution of the program, the sentence ‘Write your marks: ’ will appear on
monitor with the cursor blinking at the end of the line. Here you type the marks obtained, say
you type 60, and press the ‘enter key’, the second sentence ‘You are eligible for admission.’ will
appear on the monitor. The monitor will show the following two lines.
Write your marks: 60
You are eligible for admission.
In the following program applications of some more relational operators are illustrated.

PROGRAM 5.2 – Illustrates a multiple if (expressions)

#include <iostream>
using namespace std;
int main()
{ int n , m ,p;
cout<<“Enter three integers.”; cin>>m>>n>>p;
if(n == m) // If n is equal to m Display the following.
cout<<“n and m are equal”<<endl;
if(m !=n) // If m is not equal to n, display the following
cout << “n and m are not equal. ”<<endl;
if (m >=p) // If m is greater than or equal to p.
cout <<“m is greater than p”<<endl;
if ( p>=m)
cout<< “p is greater than m.”<<endl;
return 0;
}
❖ 102 ❖ Programming with C++

The expected output is given below.

Enter three integers.30 30 40
n and m are equal
p is greater than m.
The values of m and n are equal, while p is greater than n or m. Therefore, the statements
“n and m are not equal” and “m is greater than p” have been neglected because the if conditions
preceding them are not true.
If more than one statements have to follow the if (expression), the statements should be
enclosed between curly braces { }. A sample is shown below.
if ( expression)
{ Statement 1 ;
Statement n ; }

5.4 THE if … else STATEMENT

In many of the real life situations two or more options are available. Single if…else condition
may be used when there are two choices and a chain of if …else expressions may be used for more
than two choices. The execution of an if … else expression is illustrated in Fig. 5.2 .
Start

Initializations

else Test If true

if (expression)

Execute statement Execute statement

placed after else placed after if

To other statements

Fig. 5.2: Execution of if … else statement

The code for two choices ( if …else) may be written as below.

if (expression)
statement1 ;
else
statement2 ;
You may write the statements in a fashion that it is easier to read. Program 5.1 may be
modified to Program 5.3 given below to get a response to any value of Marks that is entered.
 Selection Statements ❖ 103❖
PROGRAM 5.3 – Illustrates if … else condition

#include<iostream>
using namespace std;
int main()
{ int Marks;
cout<<“Write your marks: ”;
cin>>Marks;
if (Marks>= 50) // conditional statement
cout << “You are eligible admission.”<< endl;

else
cout << “You are not eligible for admission.” <<endl;
return 0;
}

On clicking for execution the user is asked to enter his /her marks. Say the marks entered
are 46, the following will be displayed on the monitor.
Write your marks: 46
You are not eligible for admission.

TEST FOR DIVISIBILITY

Suppose you want the program to test whether an integer number say n is divisible by another
integer number say m. For this we use the modulus operator (%) which gives the remainder when
one integer is divided by another integer. The code is illustrated below.
if( n % m)
cout<< n <<“ is not divisible by ”<< m << endl;
else
cout<< n <<“ is divisible by ”<< m << endl;

The above mentioned if (expression) is true if there is a remainder. Therefore, if it is true then
n is not divisible by m. If there is no remainder, i.e. the condition is false then n is divisible by
m. It is illustrated by the following program.

PROGRAM 5.4 – Illustrates the test for divisibility for integers.

#include<iostream>
using namespace std ;
int main()
{
int n, m ;
cout << “Enter two integer numbers: ” ;
cin>>n >>m ;
❖ 104 ❖ Programming with C++

if (n % m)
cout<<“Number”<<n<< “ is not divisible by”<<m<<endl ;
else
cout<< “Number “<<n<<“ is divisible by ”<< m<<endl ;
return 0 ;
}

On running the program the following statement will appear on the monitor.
Enter two integer numbers:
The user is prompted to enter two numbers. At the end of line the user will find a blinking
cursor. It is here that the user has to type two numbers. First type one number say 66, give space
and type the second number say 3. Then press the enter-key. Unless you press the enter-key the
program will not proceed further. As you press enter- key the following sentences will appear on
monitor.
Enter two integer numbers: 66 3
Number 66 is divisible by 3

5.5 CONDITIONAL SELECTION OPERATOR ( ? : )

If there are two options to choose from, the conditional selection operator ( ? : ) may be used
in place of if … else . The code is quite compact and convenient. It is illustrated below.
condition ? statement1 : statement2
The above code means that if the condition evaluates true then statement1 will be carried out,
otherwise the statement2 will be carried out. Note that the two statements are separated by
colon (:). Suppose you give the name ‘max’ to the greater of the two numbers m and n, then
for selecting the greater of the two the code may be written as below.
m>n ? max = m : max = n ;
That is if (m>n) is true then the greater of the two numbers is m or max = m and if it is
not true then the greater of the two is n or max = n. Its use is illustrated in the following program.

PROGRAM 5.5 – Illustrates conditional selection operator

#include<iostream>
using namespace std;
int main()
{ int A, B, C, D ,max, Max;

cout<< “Write four whole numbers : “ ; cin>>A>>B>>C>>D ;

cout<<“Maximum of first two numbers = “ << (A > B ? A:B) << endl;
max = ((A > B ? A : B) > C ? (A > B ? A : B) : C);
cout <<“Maximum of first three numbers =” << max << endl;
 Selection Statements ❖ 105❖
Max = ((A> B ?A : B) > (C>D ? C:D)?(A>B ? A:B):(C>D ? C:D ));
cout<< “Maximum of four numbers = “ << Max << endl;
return 0;
}

The expected output is given below. The selection operator has been used to find maximum
of two, three and four numbers. The numbers entered are 60, 45, 90 and 10. The output is self
explanatory.
Write four whole numbers : 60 45 90 10
Maximum of first two numbers = 60
Maximum of first three numbers =90
Maximum of four numbers = 90

The following program gives a different application of the conditional selection operator.

PROGRAM 5.6 – Program illustrates a choice between two functions by ? : operator.

#include<iostream>
using namespace std ;
void MS1 (){cout<<“I will be the first.\n”;}
void MS2 (){cout<<“You will be the first.\n”;}
int main()
{
int mymarks,yourmarks;

cout<<“Write mymarks and yourmarks ”;

cin>>mymarks >>yourmarks;

mymarks>yourmarks ? MS1(): MS2() ;

return 0 ;
}

The expected output is given below.

Write mymarks and youmarks 92 88
I will be the first.

5.6 THE if … else CHAINS

More than one choices may be managed by a chain of if . . . . else expressions. The following
program selects the name of the 4th day of the week out of 7 choices.
❖ 106 ❖ Programming with C++
PROGRAM 5.7 – The program illustrates if . . . . else chain
#include<iostream>
using namespace std;

int main()
{ int m ;
cout<<“Name the day which is day number ”;
cin>>m ;

if (m == 1) cout<<“It is Monday”<<endl;
else
if (m == 2) cout<<“It is Tuesday”<<endl;
else
if (m == 3) cout<<“It is Wednesday”<<endl;
else
if (m == 4) cout<<“It is Thursday”<<endl;

else
if (m == 5) cout<<“It is Friday”<<endl;

else
if (m == 6) cout<<“It is Saturday”<<endl;

else cout<<“It is Sunday”<<endl;

return 0;
}

The expected output is given below.

Name the day which is day number 4
It is Thursday
Suppose you are preparing grade cards and you wish that on entering the marks the computer
should also print an appropriate quotation depending upon the marks. The following program
gives an illustration.
PROGRAM 5.8 – Another illustration of if … else chain
#include<iostream>
using namespace std;
int main()
{
int Marks;
cout << “Enter the marks: ”;
cin>>Marks ;

if (Marks >= 90)

 Selection Statements ❖ 107❖
cout<<“Excellent, keep it up”;
else
if (Marks >= 80 )
cout<<“Congratulations, you have got A grade.”<<endl;
else
if (Marks >= 60 )
cout<<“Very good, You have secured first class. ”<<endl;

else
if (Marks >= 40 )
cout<< “You have passed. You need to work hard.”<<endl;
/* The program will fail if the order in which the statements are written above is
altered.*/
return 0;
}

The expected output is given below.

Enter the marks: 75
Very good, You have secured first class.
The above program may fail to write correct quotation if the order in which the above
statements are written is changed. However, if compound Boolean conditions are written, the
program will deliver correct quotation irrespective of where the relevant statement is written.

5.7 SELECTION EXPRESSIONS WITH LOGIC OPERATORS

For compound comparative conditions such as ‘if number n is greater or equal to 80 and less than
or equal to 150’, etc. can be written with the help of Boolean logical operators which are given
below. The output of Boolean operators are either true or false. When two statements are
compared, either the condition is true which returns 1 to system or it is false which returns 0
to the system. With these return values the system gets to know whether the condition is fulfilled.
The three Boolean operators are as follows.
(i) AND (ii) OR (iii) NOT
Let A and B be the two logical statements. The following table discusses the characteristics
of the three operators under different conditions.
Table 5.2 – Boolean operators

Condition Description Operator Example of code

(i) A OR B True if either A or B or both are true. | | A| |B;
(It is false only when both are false.)
(ii) A AND B True if both A and B are true. && A && B ;
(It is false when any one or both
are false)
(iii) NOT A True if A evaluates to be false, ! !A;
and false when A evaluates true.
❖ 108 ❖ Programming with C++
PROGRAM 5.9 – The program develops a truth table for the Boolean expression !A || B.

#include <iostream>
using namespace std;

int main()
{ // for details on for expression see Chapter 6 on Iteration.

for ( int A =0; A<2 ; A++)

{ bool S;
for(int B =0;B<2; B++)
{ S = !(A||B);
cout<< “A = ”<<A << “ B = ”<<B <<“ !( A||B)= ” <<S<<endl;}}
return 0;
}

The expected output is as follows.

A=0 B=0 !( A||B)= 1
A=0 B=1 !( A||B)= 0
A=1 B=0 !( A||B)= 0
A=1 B=1 !( A||B)= 0
An application of compound expression using Boolean operators is given in Program 5.10
below. In this program we determine the maximum of the three numbers entered by the user.

PROGRAM 5.10 – Application of Boolean expressions for determining minimum of three

numbers.

#include<iostream>
using namespace std;
main()
{
int a,b,c;
cout<<“ Write three numbers: ” ;
cin>>a>>b>>c;
if (a>c && b>c)
cout <<“The minimum of the three numbers is” <<c<<endl;
else
if (c>b && a>b)

cout <<“The minimum of the three numbers is ” <<b<<endl;

else
cout <<“The minimum of the three numbers is ” <<a<<endl;
return 0;}
 Selection Statements ❖ 109❖
The program is run and three numbers 42, 25 and 75 are entered. The output of the program
is given below.
Write three numbers: 42 25 78
The minimum of the three numbers is 25
In the following program the Program 5.8 is recoded with Boolean conditions so that the
order of writing the statements does not affect the choice.

PROGRAM 5.11 – Illustrates Boolean conditions in if … else chain

#include<iostream>
using namespace std;
int main()
{ int Marks;
cout << “Enter the marks: ”;
cin>>Marks ;

if(Marks >= 80 && Marks<90)

cout<<“Congratulations, you have got A grade.”<<endl;

else

if (Marks >= 90)

cout<<“Excellent, keep it up”<<endl;
else
if (Marks >= 60 && Marks<80)
cout<<“Very good , You have secured first class. ”<<endl;

else

if (Marks >= 40 && Marks<60)

cout<< “You have passed. You need to work hard.”<<endl;
return 0;
}

The marks entered are 95 and the expected output is given below.
Enter the marks: 95
Excellent, keep it up

5.8 THE switch STATEMENT

When a multiple selection is desired we may use a if – else chain or switch statement. For large
❖ 110 ❖ Programming with C++
number of choices the if – else chain becomes unwieldy and confusing. A better method is the
switch expression which is illustrated below.
switch (expression)
{ case value1 : statement1;
break;
case value2: statement2;
break;
case value3 : statement3;
break;
.................
case valuen : statementn;
break;
default : statement; }

During execution of the program, the switch expression is evaluated. The expression should
evaluate to an integer value. Its value is compared with the values mentioned in different cases
mentioned under switch expression. If the value matches a value of a particular case, the
statements in that case are carried out. If no case-value matches with the value of switch expression
the program comes to the last statement which is a default statement. Provision of default
statement is optional. It is useful in case of user interactive programs where the user may enter
a wrong value by mistake. The statements contained in default case can remind the user that the
data was wrong and prompt the user to enter correct data. After each case the statement break;
is provided to get out of the switch at the end of a case which matched the key. If this is not
provided, then all the statements following the match would also be carried out which is not
desirable. The following points must be taken care of while using switch statement.
(i) The (expression) must evaluate to an integral value. Characters may also be used in the
switch expression because characters also have integer value as per ASCII code.
(ii) The choices are enclosed between a pair of curly brackets.
(iii) Each case value is followed by colon ( : ).
(iv) All the statements following a match are carried out. After the colon there may be one
or more than one statements. In case of multiple statements there is no need to put them
between curly braces.
(v) The break statement should be given at end of every case, otherwise, all the cases
following a match would be carried out.
The following figure illustrates the execution of switch statement.
 Selection Statements ❖ 111❖

start

switch (expression)

yes
case 1 match ? execute case 1 break

no match
yes
case 2 match ? execute case 2 break

no match
yes
case 3 match ? execute case 3 break

no match
yes
case n match ? execute case n break

no match

default case

end

Fig. 5.3: Execution of switch statement

PROGRAM 5.12 – The program illustrates switch statement

#include<iostream>
using namespace std;
main()
{
int m ;
cout<<“Name the day which is day number ”;
cin>>m ;

switch (m)
{case 1:cout<<“It is Monday”<< endl; break ;
case 2: cout<<“It is Tuesday”<< endl; break;
case 3: cout<<“It is Wednesday” <<endl; break;
case 4: cout<<“It is Thursday” <<endl; break;
case 5: cout<<“It is Friday”<< endl; break;
case 6: cout<<“It is Saturday”<<endl; break ;
❖ 112 ❖ Programming with C++
case 7: cout<<“It is Sunday”<<endl; break ;
default: cout <<“The number entered is not in the range.”<<endl;
}
return 0;
}

The following output is obtained when value 4 is entered form.

Name the day which is day number 4
It is Thursday
When the number 11 is entered for m we would get the following result.
Name the day which is day number 11
The number entered is not in the range.

Another example is that of online admissions. It may be desired to put the marks obtained
by the student as the only criterion for admission. In the following program a student is asked
to enter his/her percent marks and he/she gets the advice as to which course he/she is eligible for
admission. Six choices are provided for marks, i.e. (i) below 45, (ii) 45 and above, (iii) 60 and
above, (iv) 75 and above and (v) 90 and above. Since marks may be any number such as 46, 81,
etc., we carry out integer division of the marks by 15 and it gives an integer value less than 7.
Thus if marks are 78 the integer division by 15 will give 5 and if marks are 55 the integer
division would give 3, and so on. Also note that several statements are inserted in each case without
enclosing them in curly braces { }. Also note that each case ends with the statement break;. If
break; statement is not provided then all statements following a match, including the cases after
the match, would also be carried out which is obviously not desirable.

PROGRAM 5.13 – Another illustration of switch statement.

#include<iostream>
using namespace std;
int main()
{ int Marks;
cout << “Enter your percent marks ”;
cin >> Marks ;
switch ( Marks/15)
{ // for 90 to 100 the division will give 6
case 6: cout<<“You can get admission in all branches”<<endl;
cout<< “Finish the admission formalities by week end”<<endl;
break;
case 5 : // for marks 75 to 89
cout<<“You can get admission in all branches except the science
courses.”<<endl;
cout<< “Finish admission formalities before month end.”<<endl;
break ;
case 4: // for marks 60 to 74
 Selection Statements ❖ 113❖
cout <<“You are not eligible for Science and economics.”<<endl;
cout<< “Choose other courses by tomorrow.”<<endl;
break;
case 3: // for marks 45 to 59
cout<<“You are eligible for English and Hindi courses only.”<< endl;
cout<<“Complete admission formalities soon.”<<endl;
break ;

case 2: // for marks 30 to 44

cout<<“Sorry, Not eligible here.”<<endl;
cout<< “Better try correspondence courses.”<<endl; break;

default: cout << “Sorry, Marks entered are not correct” <<endl; }
// default case
return 0; }

The expected output on entering marks 63 is given below.

Enter your percent marks 63
You are not eligible for Science and economics.
Choose other courses by tomorrow.

EXERCISES
1. What is wrong with the following code?
if(i> 4);
n = n + I;
2. What is wrong with the following code?
if (n % m)
cout<< “n is divisible by m”;
3. Give an example of switch statement.
4. What are the Boolean logic operators?
5. Write a Boolean if( expression) for integer i to have values between 1 to 5 and between 20 to
25.
6. What is conditional selection operator? Make a small program to illustrate its application.
7. Show by constructing a truth table that the following Boolean conditions are equivalent.
!(A||B) and !A&&!B
Answer:

The truth table is constructed below. For all values of A and B it is shown that !(A||B) and !A&&!B
are equivalent.
❖ 114 ❖ Programming with C++

Truth Table

A B !A !B !(A||B) !A&&!B

0 0 1 1 1 1

0 1 1 0 0 0

1 0 0 1 0 0

1 1 0 0 0 0

8. Evaluate the two expressions !( A&&B ) and (!A||!B) for all values of A and B ( 0 and 1) , and
show that they are equivalent.
Answer: The following table shows that !( A&&B ) and (!A||!B) are equivalent.
Truth Table

A B !A !B !A||!B !(A&&B)

0 0 1 1 1 1

0 1 1 0 1 1

1 0 0 1 1 1

1 1 0 0 0 0

9. Use truth table to show that !A||B and A||!B are not equivalent.
Answer: From the truth table given below it is clear that the two expressions are not equivalent.

A B !A !B !A||B A||!B

0 0 1 1 1 1

0 1 1 0 1 0

1 0 0 1 0 1

1 1 0 0 1 1
 Selection Statements ❖ 115❖
10. Write a compound Boolean if( condition) for the following situations.
a. A varies from 0 to 10 but not between 4 to 6.
b. B varies between 20 to 30 and between 40 to 50.
Hint.
int A, B ;
if (A>=0 && A<4 || A > 6&& A <=10)
if (B>=20 && B<=30 || B>=40 && B<=50)
11. Make a program which asks the user to enter 4 numbers and program finds the maximum of
the four by using conditional selection operator.
12. Make a program to select a number greater of the two numbers which are entered by user of
the program using if (expression).
13. Find mistakes in the following program which finds minimum of three numbers.

#include<iostream>
using namespace std;
int main()
{
int a,b,c;
cout<<“Write three numbers: ” ;
cin>>a>>b>>c;
if(c<b && b<a )
cout<< “The minimum of the three number is ”<<c<<endl;
else
if (b<c && c <a )
cout<< “The minimum of the three number is ”<<b<<endl;
else

cout<< “The minimum of the three numbers is”<<a<<endl;

return 0 ;
}

14. Make a program which asks the user to enter three numbers and finds the greatest of the three
numbers using conditional selection operator.
15. Write a program to find the greatest of four numbers using the selection conditional operator.
16. Make changes to Program 5.11 in order to use switch statement in place of if – else chain.
17. Make a program for preparing the truth table for Q.7 above.
18. Make a program to prepare the truth table for Q. 8 above.
19. Find mistakes in the following program, correct it and run it.
#include <iostream> ;
using namespace std;
int main ()
❖ 116 ❖ Programming with C++
{int x, y, max;
cout<<“ Write two integers “;
cin >>x >>y;
x > y : max = x ? max = y;
cout<<Max <<“ is larger of the two numbers.”<<endl;
}

20. There are a number of mistakes in the following program. Locate the mistakes, correct and
run the program.
#include<iostream>
using namespace std;
int main()
{ int Marks;
cout << “ Enter your percent marks ”;
cin>>Marks ;
switch ( Marks/15)
{
case 6; cout<<“ You can get admission in all branches”<<endl;
break:
case 5 ;
cout<<“ You can get admission in all branches except the science
courses.”<<endl;
case 4;
cout <<“ You are not eligible for Science and economics courses.”<<endl;
break;
case 3;
cout<<“You are eligible in English and Hindi courses only.”<<endl; break:
case 2; // for marks 30 to 44
cout<<“Sorry, Not eligible here, try correspondence courses.”<<endl; break:
default; cout <<“Sorry, Marks are not correct”<<endl; // default case
}
return 0;
}

❍❍❍
CHAPTER

6.1 INTRODUCTION

In many programs it is required to repeatedly evaluate a statement or a whole block of statements

with increment or decrement of some data in order to arrive at a result. Problems such as sum
of numbers, calculations of numerical tables, numerical integration, automatic control, sorting
of lists etc. are some examples. Repeated evaluation of statements is called iteration and recursion.
The recursion is taken up in the chapter on functions while iteration which is also called looping
is the subject matter of this chapter. In many programming languages goto statement is extensively
used to go back and forth in a program. It is also used for making a loop. In C++ along with
goto there are other better options such as while, do….. while, for etc. First let us start with the
while statement.

6.2 THE while STATEMENT

In a program, the while statement or loop is incorporated as illustrated below.

while (conditional expression)
statement;
The above code means that as long as the conditional expression placed in parentheses
evaluates true, i.e. it is more than zero, the statement following this will be carried out. In the
while expression generally there is a variable whose limiting value is set to stop the loop.
However, there are endless while loops as well, which will evaluate the following statement
infinite number of times. Such loops have to be stopped with statements like break; after a
variable attains a limiting value. The listing of Program 6.1 given below carries out addition of
numbers, say from 1 to n. In the program the variables n , i and Sum are declared as integers.
The number n is initialized as 10, Sum as 0 and i as 0. The while (expression) is used in the
program as illustrated below.
while (i <= n)
Sum +=i++;
This above code means that as long as i is less than or equal to n, evaluate the following
statement which is Sum +=i++;. As you know from previous chapter this statement means that
add the value of i to value of Sum and assign this value to Sum and then increase i by 1 and again
❖ 118 ❖ Programming with C++

repeat the process. Thus the process is repeated again and again till i = 10. So Sum will become
equal to 0 + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 + 10 = 55. After the loop is over, the computer
moves to the next statement which in this program is the output statement.

PROGRAM 6.1 – Illustrates while loop for determining sum of numbers from 1 to 10.

#include <iostream>
using namespace std ;
int main()
{int n = 10, i= 0, Sum = 0;
while (i <= n)
Sum += i++;
cout << “Sum of integers from 1 to 10 = ” << Sum <<endl ;
return 0;
}

The expected output of the above program is given below.

Sum of integers from 1 to 10 = 55
The statement given in output is printed only at the end of loop, because the output
statement is not included in the loop.
Entry to while loop

Test If expression not true

(expression) Exit from loop

If expression true
Loop
Process next
statement

Fig. 6.1: The execution of a while loop

In Program 6.1 you must have noticed that the limit on the maximum value of i and its
increment each time put a limit on the number of times the loop will run. If there is no limit
on the maximum value of the variable i the loop will never end. One such condition may be
while(i>=0) followed by increment in i which starts from 0. Another example of endless loop
is while (1), the condition most commonly used for having an endless loop is given below.
while (true)
statement;
The word true is a keyword of C++ and its value is 1. In such cases it is necessary to
incorporate a break statement to get out of the loop. The different processes that take place in
the implementation of a while loop are illustrated graphically in Fig.(6.1).
 Iteration ❖ 119❖

If more than one statements of code are desired to be included in the loop they should be
enclosed between a pair of curly braces { } as given below.
while (expression)
{ statement1 ;
……………
statementn ;}
In this way the whole block of statements from statement1 to statement n becomes part of
the loop. This is illustrated in the listing of Program 6.2 in which it is desired to determine the
square roots of numbers 10, 20, 30 and 40, etc. and display them.

PROGRAM 6.2 – Illustrates while loop for determining square roots of numbers.

#include <iostream>
#include <cmath> /*Header file cmath required for square root.*/
using namespace std;
int main()
{int n = 10,N ,i=0;
while( i<5 ) // Note that there is no semi-colon at the end.
{N = i*n;
cout <<“Number = “ << N <<“\tSquare root = “<<sqrt(N)<<endl;
i++;}
return 0;
}

The expected output from the program is given below.

Number = 0 Square root = 0
Number = 10 Square root = 3.16228
Number = 20 Square root = 4.47214
Number = 30 Square root = 5.47723
Number = 40 Square root = 6.32456

In the above program the header file <cmath> has been included because we wish to calculate
square root of a number. The square root is evaluated by function sqrt() which belongs to
<cmath> header file. By using curly brackets more than one lines of code following the
while(expression) have been included in the loop. So we get output on every iteration. If the
braces {} after while (i<5) are not used the output statement would be carried out after the loop
is over, i.e. the following would be the output.
Number = 40 Square root = 6.32456

6.3 THE NESTED while STATEMENTS

When more than one parameters such as i and j are to be varied in a program and it is desired
❖ 120 ❖ Programming with C++

that for each value of i, the j be varied over its range of values, such conditions are called nested
conditions. Program 6.3 given below illustrates the use of while condition for such a case. Such
cases generally involve more than one statement of code. The i loop is the outer loop and j loop
is the inner loop. The manner in which code may be written is illustrated below.
while (int i < n)
{ statements ;
while (int j < m)
{ statements ;
statement ; } }

PROGRAM 6.3 – The program illustrates nested while loops.

#include <iostream>
using namespace std;
//The program illustrates nested while loop
void main()
{
int x =0,i=0 ;
cout<<“i\tj\tx\ty” <<endl;

while (i<=2)
{ int j = 0; // outer while loop

while (j<=2)
{ x+=(i+j); // inner while loop
int y = x*x;

cout <<i<<“\t”<<j<<“\t”<<x <<“\t”<< y<<endl;

j=j+1; }
i =i+1;}
}

The output is given as under. For each value of i the j has been varied over its range of values (0
to 2).
i j x y
0 0 0 0
0 1 1 1
0 2 3 9
1 0 4 16
1 1 6 36
1 2 9 81
 Iteration ❖ 121❖
2 0 11 121
2 1 14 196
2 2 18 324

6.4 COMPOUND CONDITIONS IN A while (expression)

For a condition such as while ( i>=2, j<= 4) the i should have initial value of 2 or more so that
the loop can start. If you put the initial value of i as 0 or 1 the loop will not start. In this while
expression the termination of the loop is controlled by condition j<=4.

PROGRAM 6.4 – The program illustrates compound condition in while (expression).

#include <iostream>
using namespace std;
int main()
{
int x =0,i=2,y;
cout<<“\ti \tx \ty” <<endl;
while ( i>=2,i<=4 ) // compound while expression
{
x = 2*i;
y = x*x;
cout <<i<< “\t”<<x <<“\t”<< y<<endl;
i =i+1; }
return 0;
}

The expected output is given below.

i x y
2 4 16
3 6 36
4 8 64

More than one conditional expressions with different variables separated by comma (,) may
be put in while expression as illustrated below.
while ( i>=2, j<=5, k <= 6)
In such cases the loop is terminated by the last condition in the expression. In the above
mentioned case, the value of k controls the termination of loop. However, the output may as
well depend on the compiler you are using. Program 6.5 provides an output.

PROGRAM 6.5 – Another example of compound condition in while expression.

#include <iostream>
using namespace std;
❖ 122 ❖ Programming with C++

int main()

{
int x =0,i=2,j=0,k = 0,y;
cout<<“i\tj\tk\tx\ty” <<endl;

while (i>=2, i<=4,j<=5 ,k <=6 ) // compound while condition

{
x =(i+j+k);

y = x*x;
cout <<i<<“\t”<<j<<“\t” <<k <<“\t”<<x <<“\t”<< y<<endl;
j++;
i ++;
k ++;
}
return 0;
}

The expected output, given below, shows that the intermediate conditions, i.e. i <= 4 and
j <= 5 are ineffective. The while expression is evaluated in every iteration. In case i <2 the loop
will not start. Once started, the loop will stop when the last expression in the while (expression)
evaluates false. For compliance of all the conditions it would be necessary to have the conditions
connected by Boolean AND operator.
i j k x y
2 0 0 2 4
3 1 1 5 25
4 2 2 8 64
5 3 3 11 121
6 4 4 14 196
7 5 5 17 289
8 6 6 20 400

COMPOUND BOOLEAN CONDITION IN WHILE (EXPRESSION)

Generally in most of the programs only one condition in while expression is used, however, there
is no bar to use compound conditions. This has been demonstrated in Program 6.5. If the
expressions are simply separated by comma (,) the intermediate conditions may be overlooked,
however, if they are connected by Boolean && operator the loop would end with any one
variable reaching the limit. The Program 6.5 is modified and run again with Boolean condition
as given below.
while (j<5&&i>=3&&i<=10)
 Iteration ❖ 123❖

With this expression the loop will end when any one condition is violated.

PROGRAM 6.6 – The program illustrates while loop with compound Boolean expression.

#include <iostream>
using namespace std;
int main()

{int x =0,i=3,j=0,y;
cout<<“\ti\tj\tx\ty” <<endl;

while (j< 5 && i>=3 && i<= 10)

{ x+=(i+j);
y = x*x;

cout <<“\t”<<i<<“\t”<<j<<“\t”<<x <<“\t”<< y<<endl;

j++;
i++;
}
return 0;
}

From output given below it is clear that i has been varied from 3 onward while due to the
condition on j the loop is terminated at j = 4. If Boolean operators were not put and the variables
were simply separated by (,) the loop would have gone to i = 10 and j would have gone to 7
beyond the condition j<5. The output of the program is given below.
i j x y
3 0 3 9
4 1 8 64
5 2 15 225
6 3 24 576
7 4 35 1225

If, however, the above while expression is put as while (j> = 0 || i> = 3 && i<= 10), and
the initial value of i may be less than 3, the condition will evaluate to be true and the program
would run from j = 0 to i = 10.

6.5 THE do….while LOOP

The use of do…while loop is similar to while loop except for the fact that the while
condition is evaluated at the end of program, therefore, at least one computation would be carried
out even if the while condition turns out false. The following figure illustrates the execution of
do … while loop.
❖ 124 ❖ Programming with C++

Star t

Initialization &
other processes
Entry to do -- while loop
Process
statements

Test If expression not true

Loop (expression) Exit from loop

If expression true

Fig. 6.2: Execution of do — while loop

The following program illustrates the application of do … while loop.

PROGRAM 6.7 – The program illustrates do --- while loop.

#include <iostream>
using namespace std;
void main()
{
int x =0,i =0 , y;
cout<<“i\tx\ty” <<endl;

do // start of do – – while loop

{
x =2*i;
y = x*x;

cout <<“”<<i<< “\t”<<x <<“\t”<< y<<endl;

i++;}
while ( i<=3 ); //Test condition of do –– while loop
}

The expected output is given below.

i x y
0 0 0
1 2 4
2 4 16
3 6 36
 Iteration ❖ 125❖
The same program gives following output when i is initialised as i = 5 instead of i = 0. Even
though the condition i <=3 is violated, the program has been executed for one time and gives
the following output.
i x y
5 10 100
FACTORIAL OF A NUMBER
In the following program an application of do…while loop has been made to evaluate the factorial
of positive numbers. The break command has been used when the factorial exceeds a certain
value.
Factorial (n) = 1× 2 × 3 × 4 × 5 … × ( n–1) × n
Or it may also be written as
Factorial (n) = n × Factorial (n-1)
which is a recursive form of function. This aspect is discussed in Chapter 7. In the following
program the first form is programmed.
PROGRAM 6.8 – Illustrates calculation of factorial numbers with do---while loop.
#include <iostream>
using namespace std;

int main()
{
int x =0,i=0,factorial=1, limit = 500;
cout<<“i”<<“\tx”<<“\t” << “Factorial x” <<endl;

do
{ i =i+1;
x +=1;
factorial *= x;
cout <<i<<“\t”<<x <<“\t”<< factorial<<endl;
}
while (factorial<limit );

return 0;
}

The output is given below. The output shows that maximum limit of 500 has been crossed.
This is because the while expression is tested at the end of the loop, so when the condition is
tested it is already beyond the specified limit.
i x Factorial x
1 1 1
2 2 2
3 3 6
4 4 24
5 5 120
6 6 720
❖ 126 ❖ Programming with C++

6.6 ENDLESS while LOOPS

To get out of an endless loop such as while(true) we can make use of statement break; However,
if you also want to get out of the program as well, you may use exit() function. With break
statement one gets out of the loop while with exit() function one gets out of the program. The
argument for exit() function is 0 or any integer. For some compilers 0 conveys successful
termination. Thus it may be coded as
exit (0);
The function is placed after the statement where it is desired to get out of the program.
PROGRAM 6.9 – Illustrates use of exit(0) to terminate the endless loop and the program

#include <iostream>
using namespace std;

int main( )
{cout<<“i\tx\ty” <<endl;
int x =0,i=0 ,y;
while (true) // Endless while loop
{
x = i*i ;
y = x*x;
if (y>300)
exit(0); // 0 or any integer value can be argument
cout <<i<<“\t”<<x<<“\t”<< y<<endl;
i++;
}
return 0;
}

The expected output is given below.

i x y
0 0 0
1 1 1
2 4 16
3 9 81
4 16 256
In the above program while (true) has been used, which is an endless loop. The loop is
terminated by putting a limit on the value of one of the altering variable as the condition for
exit. Such loops may also be terminated by using the statement break ; in the code. This is
 Iteration ❖ 127❖

illustrated in the following program. You should note that the break statement ends with a
semicolon.

PROGRAM 6.10 – The program illustrates endless while loop with break statement.

#include <iostream>
using namespace std;
int main(){
int x =0,i =1,y;
cout<<“i”<<“\tx”<<“\t” << “y” <<endl;

while (true) //Endless loop

{ x +=i;
y = x*x ;
cout <<i<<“\t”<<x <<“\t”<< y<<endl ;
i++;
if(i>= 6)
break; // break statement
}
return 0;
}

The expected output is as under.

i x y
1 1 1
2 3 9
3 6 36
4 10 100
5 15 225

6.7 THE for LOOP

The for loop is coded as given below.

for (initial value of variable ; conditional expression ; mode of increment /decrement)
statement ;
There are three expressions inside the brackets separated by two semicolons. In the first
expression the variable controlling the loop is initialised, i.e. its initial value is specified, in the
second expression its final or limiting value is given, i.e. the value of the variable at which the
loop should terminate, in the third expression the mode of increment/decrement is specified. The
working of a for loop is illustrated in Fig. 6.3.
❖ 128 ❖ Programming with C++
Entry to for loop

Initialization

Test for If expression not true

(expression) Exit from loop

If expression true
Loop
Execute next
statement

Increment/
decrement

Fig. 6.3: Execution of a for loop

A for loop is illustrated below. In the following expression the variable i controls the loop.
for (int i = 0 ; i< = 10 ; i++)
statement ;
If more than one statements are to be included in the loop they should be enclosed between
the curly braces {} as illustrated below.
for( int j = 20 ; j>=0 ; j–)
{ statement1;
—————
statementn ; }
The program below carries out sum of numbers 1 to 10 with the help of for loop. The
output is displayed at the end of loop.
PROGRAM 6.11 – Program illustrates for loop for sum of numbers from 1 to 10.
#include<iostream>
using namespace std;
main()
{int n = 10;
sum=0 ;

for (int i=0;i<=n; i++) // beginning of loop statement

sum += i; // end of loop statements

cout<<“sum = ”<<sum<<endl;
return 0;
}
 Iteration ❖ 129❖
In the above program you would observe that loop starts from the line for (int i=0;i<=n;i++)
and ends at the statement sum + = i;. For the next iteration, it goes back to the loop expression
and repeats the process. So in every iteration the value of i is added to variable sum. At the end
of looping process the sum is 55 which is displayed by output statement of the program. The
expected output is given below.
sum = 55

ENDLESS FOR LOOP

You should note that inclusion of the expressions in the brackets are optional but inclusion of the
two semicolons is a must. We may as well write the above for loop as
int i = 0;
for (; i<10; )
i++; // See program 6.26

We may write an endless for loop as below.

for (; ; )
In the above declaration there are no expressions except the two semicolons. This is an
endless for loop and as already discussed above an endless loop may be stopped by using the
break; statement. See the following program for an illustration.

PROGRAM 6.12 – Illustrates an endless for loop.

#include<iostream>
using namespace std;
int main()
{ int i =0,sum=0 ;

for(;;) //endless for loop

{ sum += i;
if (sum >25) break; // break included to end the loop.
cout<<“i= ”<<i<<“ sum = ” <<sum<<endl;
i++;}
return 0;
}

The expected output is given below.

i= 0 sum = 0
i= 1 sum = 1
i= 2 sum = 3
i= 3 sum = 6
i= 4 sum = 10
i= 5 sum = 15
i= 6 sum = 21
❖ 130 ❖ Programming with C++

6.8 COMPOUND CONDITIONS IN for (expression)

The for (expressions) may have conditions on more than one variables. In such compound
conditions the variables are separated by comma as illustrated for m and n below.
for (int n =0, int m = 0; n <=3,m <=5; n++,m++)
In the above expression the initial values of variables are separated by comma (,), the limiting
values are separated by comma and finally the modes of increments are also separated by comma.
The different variables may have different modes of increment/decrement. In Program 6.13 given
below two variables n and m are included in for loop with differing limiting values but similar
modes of increment.

PROGRAM 6.13 – Illustrates compound condition in for (expression).

#include<iostream>
using namespace std;
int main()
{int n, m ;

for (n=2,m=0;n<=8,m<=3 ;n++,m++)

cout<<“n = “<<n<<“, m = “<<m<<“, product = “ <<n*m<<endl;
return 0;
}

The expected output is as below.

n = 2, m = 0, product = 0
n = 3, m = 1, product = 3
n = 4, m = 2, product = 8
n = 5, m = 3, product = 15

In the above case the loop terminates at m<=3 and does not continue to n<=8. If the same
program is run again with the condition for(n=2,m=0;n<=3,m<=8 ;n++,m++), i.e. by
interchanging the limits the loop will run up to m<=8. The control rests with the limiting factor
which is last in the middle expression.

THE FOR LOOP WITH COMPOUND BOOLEAN EXPRESSION

If Boolean operators are used in for (expression) the above stated problem is not there, i.e. all the
conditions are interconnected by Boolean operators. This is illustrated in the following program
in which the for expression is
for (int i=0,j=0; i<= 4 &&j <= 16; i= i++, j= j+2)
In this expression the minimum limiting value is i<=4 so loop will end at this value of i,
because the limiting condition would evaluate true only up to i <=4. It is illustrated below.
 Iteration ❖ 131❖

PROGRAM 6.14 – Illustrates compound Boolean condition in for (expression).

#include <iostream>
using namespace std;
int main()
{ int Product;
for (int i=0,j=0; i<= 4 &&j <= 16; i++, j= j+2)
{ Product = i*j;
cout<<“ i = ”<<i<< “\tj = ”<<j << “\tProduct = ”<< Product<<endl; }
return 0;
}

The expected output is given below.

i = 0 j = 0 Product = 0
i = 1 j = 2 Product = 2
i = 2 j = 4 Product = 8
i = 3 j = 6 Product = 18
i = 4 j = 8 Product = 32

6.9 NESTED for LOOPS

If a function involves more than one variable and we want to evaluate it for different values of
all the variables, we will have to use a nested for loop. For sake of illustration say we have a function
with three integer variables m, n and p. If it is desired that function be evaluated for all values
of n, m and p then a nested for loop, as illustrated below, may be used.
for ( int n=0; n<= A; n++)
for ( int m=0; m<= B ; m++)
{ Statements;
----------
for ( int p =0 ; p<= C ; p++)
{ Statements;
--------- }
}
Program 6.15 illustrates the code for a nested for loop for two variables n and m. In a similar
way a program may be written with any number of nested loops.

PROGRAM 6.15 – Illustrates nested for loops.

#include<iostream>
#include<iomanip>
using namespace std;
int main()
{
❖ 132 ❖ Programming with C++
int n, m ;
for (n=1;n<=3;n++)
for(m=1; m<=2; m++)
{cout<<setw(5);
cout<<“n = “<< n<<“ m= “<<m <<“ product = “<<n*m <<endl;
}
return 0;
}

The output of the above program is given below. You will observe that for every value of n, the
variable m has been varied over its range of values, i.e. 1 and 2 .
n = 1 m= 1 product = 1
n = 1 m= 2 product = 2
n = 2 m= 1 product = 2
n = 2 m= 2 product = 4
n = 3 m= 1 product = 3
n = 3 m= 2 product = 6

6.10 GENERATION OF RANDOM NUMBERS

In many real life problems there is an element of uncertainty and require probabilistic or random
sampling technique. Examples are in quality control, queues on service stations, computer games,
etc. For treatment of such problems random numbers are required. A random number is one
whose value cannot be predicted just as the time of occurrence of an earthquake cannot be
predicted, at least, at present. It is nearly impossible to get truly random numbers, however,
pseudorandom numbers may be generated. In C++ the following two functions are used which
are provided in the header file <cstdlib>.
int rand();
void srand (unsigned seed);
The function rand() produces random numbers in the range from 0 to a maximum
number called RAND_MAX provided in the header file. If only rand() is used it would
produce same set of random numbers every time it is run. This is illustrated in Program 6.16
given below. However, if a seed number is provided to random number generator by the function
srand() it would produce a different set of random numbers. Every time rand() is run a
different seed number has to be provided to get a different set of random numbers. This is
illustrated in Program 6.17 given on the next page.

PROGRAM 6.16 – Illustrates generation of random numbers with rand() function.

#include<iostream>
#include<cstdlib>
using namespace std;
int main()
 Iteration ❖ 133❖
{int n ;
for (n=1;n<=6;n++)
cout << rand()<<“\n”;
return 0;
}

Output of the first trial: In the above program no seed number is fed and rand() function
produces the set of 6 random numbers given in the output below. If the program is run again
it will produce the same set of random numbers as illustrated in the second run of the program.
41
18467
6334
26500
19169
15724
The output of the second trial is given below. It is same as the output of first trial.
41
18467
6334
26500
19169
15724

From the above outputs it is clear that every time the random number function is called it
will produce the same set of random numbers. Such a condition is not desirable because it smells
some kind of setting. Many times, in situations such as in draw of lots it is required to assure the
audience that the random number generated are not in any way biased. In such cases, a person
from the audience may be asked to enter a seed number. This is a case where user of program
provides the seed number. Program 6.17 given below solves this problem. In the first trial run,
the seed number provided is 211and the random numbers generated are given in the output. In
the second trial the seed number is 151 and we get a different set of random numbers in the output.

PROGRAM 6.17 – Illustrates random number generation with a seed number.

#include<iostream>
#include<cstdlib>
using namespace std;

int main()
{
int n , S ;
cout<< “Write a positive seed number : ”;
cin>> S;
❖ 134 ❖ Programming with C++
srand(S);

for (n=1;n<=6;n++)
cout << rand()<<“\n”;
return 0;
}

With seed number 211, the following output is obtained.

Write a positive seed number : 211
727
14641
22258
756
22542
21137

With seed number 151 the output is as below.

Write a positive seed number : 151
531
25096
31752
31517
26260
14909

In programs such as computer games it is not possible for user to provide a seed number on
every event in the game. In order to get around such problems we may make use of function
time() defined in the header file <ctime>. The function time(Null) gives a number which
is equal to the number of seconds elapsed up till now since 1st January 1970 Greenwich time
00:00:00. This number is always changing and can provide a different seed number every time
rand() is run. The code is srand (time(Null)); .
In the following program, time has been used for the seed number. The header file <ctime>
of C++ standard library has been included. The code for using time () as seed number is as below.
long unsigned time(Null);
The program given below also displays the value of seed number.

PROGRAM 6.18 – Illustrates use of time( ) as seed number.

#include<iostream>
#include<cstdlib>
#include <ctime>
using namespace std;
main()
 Iteration ❖ 135❖
{int n ;
long unsigned seed(time( NULL )) ;
srand(seed);
cout <<“seed = “ << seed<<endl;
cout<< “The random numbers are as below.”<<endl;
for (n=1;n<=6;n++)
cout << rand()<<“\n”;
return 0;
}

The output is given below.

seed = 1183200663
The random numbers are as below.
29589
3656
1276
1828
24019
21122

RANDOM NUMBERS IN A SPECIFIED RANGE

For obtaining random numbers between zero and a limiting high value say High, we may take
the modulus of random numbers generated by rand() with (High +1). However, if it is desired
to obtain random number between two limits say High and Low we may use the following
code.
Randnumber = rand() %(High – Low +1) + Low;

PROGRAM 6.19 – Illustrates generation of random numbers in specified range.

#include<iostream>
#include<cstdlib>
#include<ctime>
using namespace std;
int main()
{
long Randnumber ;
long unsigned seed(time( NULL )) ;
srand(seed);
cout<< “seed = “<<seed<<endl;
cout<< “Random numbers in two digits (less than 100)”<<endl;
for (int n=1;n<=10;n++)
❖ 136 ❖ Programming with C++
{ Randnumber = rand();
cout << Randnumber %100<<“ “;
}
cout << “\n”;
cout<< “Random numbers for dice play (High = 6, Low = 1).\n”;
for (int k=1;k<=10;k++)
{ Randnumber = rand();
cout << Randnumber%6 + 1<<“ “; //for High = 6, Low = 1.
}
return 0;
}

The expected output is given below.

seed = 1225696985
Random numbers in two digits (less than 100)
15 19 36 10 22 84 31 59 40 13
Random numbers for dice play (High = 6, Low = 1).
6 2 3 5 6 4 4 3 4 4

6.11 THE goto STATEMENT

The code goto is used in many programming languages for moving back and forth in the program.
In C++ the code lines are not numbered as is the case in some other programming languages.
Therefore, for using goto statement one needs to put in a label. At the starting point, i.e. point
from where the jump is to be executed we simply code as goto name_of_label as illustrated below.
statements;
goto name_of_label ; //code at point of jump
——————
statements;
name_of_label : //code at destination of jump.
statements;

At the destination, the label name is put at the start of line and it has to be followed by colon
( : ) as illustrated above. Any other statement is put on right side of colon or in next line. In the
following program we have put the label name as Again. So at the point of jump the statement
is written as goto Again;. At the destination it is Again: . The following program illustrates the
application of goto statement for iteration.
 Iteration ❖ 137❖
PROGRAM 6.20 – Illustrates use of goto statement for going back in program.

#include <iostream>
using namespace std;
int main()
{
int n , m ;
cout <<“Write and enter two integers: “ ; cin>> n>> m;
cout << “You have written the numbers as n= “<<n <<“ and m = “<<m <<endl;

Again : //The label Again, see colon at end.

if(n<m)
n++;
else
m++;
if(n == m)
cout<<“Now m = ” << m<<“ and n = ” << n<<“\n”;
else
goto Again; // Jump back to Again making a loop
return 0;
}

The expected output of the above program is given below.

Write and enter two integers: 40 60
You have written the numbers as n= 40 and m = 60
Now m = 60 and n = 60

In the following program goto command has been used for moving forward in the program.

PROGRAM 6.21 – The program illustrates goto command with a while loop.
#include <iostream>
using namespace std;
int main()
{
cout<<“i\tx\ty” <<endl;
int x =0,i= 0 ,y;
while (true)
{ i =i+1;
x = 2* i ;
y = x*x;
if (y>100)
goto End; // ‘End’ is the name of label.
cout <<i<<“\t”<<x <<“\t”<< y<<endl;
}
End:
return 0 ; }
❖ 138 ❖ Programming with C++

The expected output is given below.

i x y
1 2 4
2 4 16
3 6 36
4 8 64
5 10 100

6.12 THE continue STATEMENT

The continue command may be used to bring the control back to the top of a loop before all
the statements in the program are executed. The following program illustrates the application
of continue and goto commands. In the program an integer 20 is decreased by 1 in each iteration
till a number divisible by 7 is reached. On getting the number the goto statement takes the
program to the end.

PROGRAM 6.22 – The program illustrates the use of continue and goto commands.

#include <iostream>
using namespace std;

int main()
{
int n =7 , m=20 ;
while(n<m)
{m= m–1;
if (m% n ==0)
goto end;
else
continue; }
end: cout<<“m = ” << m<<“ and n = ” << n<<“\n”;
return 0; }

The expected output obtained on running the program is given below.

m = 14 and n = 7

6.13 INPUT WITH A SENTINEL

Sometimes it is not known as to how many entries will be there or the entries may be limited
but you do not wish to count them. For such cases you may use an endless loop for input of
data using cin function. You may also incorporate a conditional statement that if an entry value
 Iteration ❖ 139❖
is equal to a specified value which is not likely to be there in the usual entries, the program should
get out of loop and process the data already entered. This is called input with a sentinel.
The following program illustrates an application of input with a sentinel.
PROGRAM 6.23 – Illustrates input with a sentinel.
#include<iostream>
using namespace std;
main()
{
int Marks , Entry_no = 0 ;
double Av_Marks, total = 0.0;
while (1)
{ Entry_no++;
cout<< “Entry_no = ” <<Entry_no <<endl;
cout<<“Enter the marks = ”; cin >> Marks;

if (Marks >100) {cout<<“last entry not counted \n”; break;}

total = total + Marks;
Av_Marks = total/Entry_no;
}
cout << “Total Entries = “ <<Entry_no <<endl;
cout << “Average Marks = “ << Av_Marks << endl;
return 0 ;
}

In this program it is taken that the maximum marks are not going to be more than 100.
So the condition for break has been put as if (Marks > 100) break; After getting out of the loop
the program calculates the average marks obtained by students. The output of the above program
is given below.
Entry_no = 1
Enter the marks = 50
Entry_no = 2
Enter the marks = 80
Entry_no = 3
Enter the marks = 90
Entry_no = 4
Enter the marks = 60
Entry_no = 5
Enter the marks = 44
Entry_no = 6
Enter the marks = 110
last entry not counted
Total Entries = 5
Average Marks = 64.8
❖ 140 ❖ Programming with C++

EXERCISES
1. How do you write a for loop expression?
2. Give a sample code for endless for loop.
3. Make a program using endless for loop and use break; statement to get out of it.
4. Make a small program to illustrate an endless while loop and use exit () function to end the
program.
5. Write a program using while loop to make a multiplication table of 1.25 from 1×1.25 to 10×1.25.
6. What is the difference between while loop and do … while loop?
7. Make a program using for loop to display square roots of numbers from 1 to 5.
8. Write a program to convert inches into millimetres and millimetres into inches.
9. The function 2y + 3xy –6zx + 4z is to be evaluated for x = 1 to 4, for y = 2 to 5 and z = 3 to
6. Write a program with nested for loops to achieve this.
10. Write an endless for loop to read a number of integers. It should exit when 000 is entered.
The program should display total number of entries and average value of the entered data.
11. Make a C++ program to generate random numbers with seed number provided by user.
12. Make a program with for loop to generate 6 random numbers less than 1000 by using function
time() for generating the seed number.
Answer: Type and run the following program to get a set of random numbers less than 1000.

PROGRAM 6.24 – Illustrates generation of three digit random numbers.

#include<iostream>
#include<cstdlib>
#include<ctime>
using namespace std;
int main()
{
long Randnumber ;
cout<< “Random Numbers are”<<endl;
long unsigned seed= time( NULL ) ;
srand(seed);
for (int n=1;n<=6;n++)
{Randnumber = rand()%1000;
cout << Randnumber<<“\n”;
}
return 0;
}
 Iteration ❖ 141❖

The following output is obtained by running the program.

Random Numbers are
411
809
767
51
671
123
13. Find mistakes in the following program. Correct it and run it.

#include <iostream>
using namespace std;
void main()
{
int x = 0;
do;
{
cout<< “\tcube of “<< x <<“ = “<<x*x*x <<endl;
x +=1;
}
while(x<=4)
return 0;
}

14. Write an iteration program with endless while loop. Make use of exit() function to get out
of it.
15. Make a program with do—while loop to calculate the square roots of numbers from 1 to 10.
16. Write a program with endless while statement and make use of break statement to terminate
the loop.
17. Make a program to illustrate the use of goto statement.
18. Make program to illustrate the use of continue statement.
19. Make a program to evaluate the following function for x = 4 to 8, y = 2 to 3 and z = 5.
4x3 z + 5 y z + 10 x y z
20. Construct a program with for loop to evaluate a truth table for the Boolean expression. !(A||B)
Answer:
#include <iostream>
using namespace std;
int main()
❖ 142 ❖ Programming with C++
{ int A , B ;
for ( A =0; A<2 ; A++)
{for(B =0;B<2; B++)
cout<< “A = ”<<A << “ B = ”<<B << “ !( A||B)= ” << !(A||B)<<endl;}
return 0;
}

21. Make a program for integration of x in the limits specified by user.

Answer:
For integration of a function f(x) between the limits a and b, we follow the well known trapezoidal
rule as illustrated below.
b 1 1 nh3 ||
∫a f (x) dx = h ( f 0 + f1 + f 2 + f3 + … + f n–1 + f n ) – f (ξ)
2 2 12
Where
h = (b – a)/ n
f0 = f (a)
fn = f (b)
fk = f (a + k × h)
The last term goes to zero as n goes to ¥. The method is illustrated in the following
program.
PROGRAM 6.25 – Illustrates calculation of integral of x between the limits specified
by user.
#include <iostream>
#include <cmath>
using namespace std;
int main () {
double x , Lower_Limit , Upper_Limit , Integral , h, F;
int n = 100;
cout<<“Enter the limits of integral :\n” ;
cout<< “Upper_Limit = “; cin >> Upper_Limit ;
cout<< “Lower_Limit = “ ; cin>> Lower_Limit;
h = (Upper_Limit – Lower_Limit )/n;

Integral = 0.5 * (Lower_Limit + Upper_Limit)*h;

for ( int i =1; i<n; i++)
{F = (Lower_Limit + i*h)*h ;
Integral += F; };

cout << “Integral of x = “ << Integral <<endl;

return 0 ;
}
 Iteration ❖ 143❖
The expected output is given below. The user has put in the limits as 8 and 2 so the result is 30.
Enter the limits of integral :
Upper_Limit = 8.0
Lower_Limit = 2.0
Integral of x = 30
22. Make a program to integrate xn in the range for x = a to x = b. Also develop a program to
integrate any polynomial in the limits a to b.
23. Make a test program to illustrate endless for loop.
24. Make a program with for loop having only middle expression.
Answer: The program is given below.

PROGRAM 6.26 – Illustrates for loop with only middle expression.

#include<iostream>
using namespace std;
main()
{int n = 10,sum=0, i=0 ;
for ( ;i<=10; )
{sum +=i;
i++;}
cout<<“sum = ”<<sum<<endl;
return 0;
}

The program runs and output is as below.

sum = 55

❍❍❍
❖ 144 ❖ Programming with C++

CHAPTER

7.1 INTRODUCTION

Generally we take a function as some formula type mathematical entity whose numerical value
has to be found if parameters are given. However, in C++ along with the formula type, a function
in the broader sense may be considered as a component of a big program. A large program may
be divided into suitable small segments and each segment may be treated and defined as a function.
These segments can be separately compiled, tested and verified. So division of a very large program
into manageable smaller functions or modules not only makes the development process easier
it also makes it possible to involve a large number of programmers in the development process.
The different functions/program segments may be developed separately by different programmers
and then combined into a single program. Such a scheme makes it easier to locate bugs.
Moreover, the maintenance of a big program also becomes easier.
The purpose of a function is to receive data, process it and return a value to the function
which has called it. The terms calling function and called function are derived from the telephone
communication. The one who rings the number is the calling person and one who receives the
call is the one called. The same terminology applies to functions. Thus the function which calls
another function is the calling function and the function which is called is the called function. The
different data variables that a function accepts for processing are called parameters of the function.
The values that represent the parameters and are passed on to the function when it is called, are
the arguments of the function. Some functions return a numeric value to the calling function.
These are called return type functions. If a function does not return a numeric value we call it
void function. A function may be void but it still may be processing integer type, floating type
or other types of data. The function is still called void. Similarly a function may not be void
but still may not have any argument to operate upon, in that case, its arguments are void.
A programmer should understand how to declare a function, define a function and call a
function. In a program a function has to be declared and defined only once. Then it may be called
as many times as required. The functions which are declared and defined by a programmer are called
user defined functions or programmer defined functions. Besides, C++ Standard Library has a large
collection of predefined functions which can also be used by a programmer. We have already dealt
with some such functions such as swap (), sqrt (), etc. We shall discuss more of these in this chapter.
Below, Fig.7.1 shows a general procedural program in which other functions are called.
 Functions ❖ 145❖

// The Main Program. Main program calls function F1. F1 calls another function F3.

# include <iostream> When a function is

Using namespace std; called the control
Function1 Prototype; shifts to the called
Function2 Prototype; function

int main ()
{ —————
Function1 (arguments) ;
Function1 (parameter list) F3 (parameter list)
{ statements; (statements;
— F3 (arguments); statements;
( ------ return value;}
statements --------
------------- statements;

statements; return value ;}

Function2 (arguments);
On return the control goes back to
calling function
Function1 (List of parameters)

{statements; }
Function
Function2 (List of parameters) definitions
{statements;
------ ;}

Fig. 7.1
Figure 7.1 shows that main program contains a number of statements and also function calls.
When function F1 is called the action already taken by computer on main program is saved and
control shifts to the called function F1. The function F1 during its execution calls another
function F3 and the action already done in F1 is saved and control shifts to the function F3.
On completion of F3 the control comes back to F1 and on completion of F1 the control comes
back to the main program. The main program proceeds from where it was left when F1 was
called.

7.2 USER DEFINED FUNCTIONS

A function has a head and a body. A complete definition of a function comprises both head and
body. The head gives the information about type of function, name of function and a list of
parameters enclosed in parentheses. The type of function is the type of data it returns. Thus if
the data returned is integer the function type is int. If the function returns a floating point
number, the type of function is float or double, etc. If the function does not return any numeric
data its type is void. However, it is possible to separate the declaration of head or prototype of
function and the definition of a function. A function body comprises statements and expressions
enclosed between a pair of braces {} and which perform the task of function. The function
definition which comprises both head and body, is illustrated below.
❖ 146 ❖ Programming with C++

type identifier (type parameter1, type parameter2, ---, type parameter n) // head of function
{statements ;} // body of function
In the above definition the first word is the type of function, it is the type of data it returns.
The second item is identifier which is the name of function. The name may be decided in the
same way as we decide the name of a variable. One has to follow the same rules as given in
Chapter 3 for identifiers for variables.
The method of declaration of parameters (third item) is equally important. The parameters
are written between a pair of parentheses ( ). The type of each parameter such as int, float , double
, char, etc., has to be mentioned individually before the name of the parameter otherwise the
compiler will take it as an integer by default. For example, suppose we want to define a function
to represent the weight of a cubical body having length, width and height in whole numbers and
density of its material in floating point number. The function definition is illustrated below.
double Weight (int Length, int Width, int Height, float Density)
{ return Length*Width*Height*Density ;}
Here the name of function is Weight. The first word double is its type because the type of
return value is double. Length, Width, Height are names of three integer type parameters
representing length, width and height respectively, and, Density is the name of a parameter which
is a floating point number and represents density of material. You should note that each parameter
is preceded by its type individually.
The body of function comprises the statements and expressions required to perform the task
of function, and are enclosed between curly braces {}. In the above definition there is only one
statement which returns the product of all the arguments. Arguments are the values of
parameters, some authors call them as actual parameters. The following points should be noted
when declaring a function head.
(i) All the parameters are enclosed in the parentheses and are separated by comma.
(ii) The type of each parameter is declared individually. If the type of a parameter is not
mentioned the compiler may take it as an integer by default.
(iii) There is no semicolon at the end of head line if it is followed by body otherwise there
is semicolon.(;) at the end of header.
Now we have to decide as to where the function definition be placed in a program. A
function cannot be defined inside another function. Therefore, the definition cannot be placed
inside the main () function. It has to be outside the main () function. In a program, function
definition may be placed before main() function, above or preferably below the names of header
files or it may be done after the end of main () function. The declaration and definition may
be separated as well, i.e. the declaration (function head followed by semicolon) or function
prototype may be placed above the main () and definition (both head and body) may be placed
below the end of the main () function. Some compilers require that function head or prototype
be given before it is invoked in the main() function. The definition may be placed after the end
of main () function. Fig. 7.2 illustrates the different placements of a function in a program. The
function prototyping is discussed after this.
 Functions ❖ 147❖
// Main program // Main program

# include <iostream> # include <iostream>

using namespace std; using namespace std;
Function Prototype; Function definition
int main () int main ()
{ { statements ;
statements ;
Function (arguments) ; Function (arguments);
———— // Function Call
return 0; } return 0 ;
}
Function definition
Fig. 7.2 (a): Prototype above the main() Fig. 7.2 (b): With function definition above main,
and definition below the main() no prototype is needed.

//Main program
# include <iostream>
using namespace std;
Function head ;
int main ()
{ ————
statements;
Function ( arguments );
————— // Function Call
return 0 ;
}
Function definition
Fig. 7.2 (c): Function head declared above
main () and function defined after main ()

Fig. 7.2 (a, b, c) – Where to place the definition of a function in a program?

7.3 FUNCTION PROTOTYPE AND RETURN STATEMENT

A function prototype comprises type and name of the function along with list of types of its
parameters enclosed in brackets. It ends with a semicolon as illustrated below.
type identifier (type-of- parameter1, type-of-parameter2, … type-of- parameter n );
The prototype conveys the following information to the compiler.
(i) The type of function, i.e. the type of return value.
(ii) Number of parameters of the function.
(iii) Type of each parameter.
(iv) The order in which the arguments would be placed when the function is called.
The method of declaring the function prototype before the main() and the function definition
(head and body) after main () function is more popular. It is illustrated in Fig.(7.2 a). In this
❖ 148 ❖ Programming with C++

arrangement you need not define all the functions right in the beginning of the program. Only
function prototypes are given at the beginning.
It is very important that in function prototype the type of each parameter is mentioned
individually and types are placed in the same order in which the corresponding arguments
would be placed when the function is called. The names of parameters may or may not be given
in the prototype, however the names are necessary in the head of function in the function
definition. For example, the prototype of function Weight may be declared as below.
double Weight ( int, int, int, double) ;
Observe that names of parameters are not given. In fact names are not required to be given
in the prototype of a function. Compiler does not take note of names in prototype. However,
if you have mentioned names the complier will not give error either. Therefore, you may as well
write the prototype of function Weight, discussed above, as
double Weight(int Length, int Width, int Height, float Density);
You should note that a function prototype ends with a semicolon (;).
For calling the function in the main() or in any other function you need to mention the
name of function followed by arguments (values of parameters) in parentheses and end the line
by a semicolon. The above function Weight may be called as below.
Weight (a, b, c, d);
Here a, b, c and d are the actual values of parameters or arguments (some authors call actual
parameters). In this instance, Length takes value a, Width takes the value b , Height takes the
value c and Density takes the value d. It is important that the values are put in the same order
as declared in the prototype, because the compiler would take them in the same order. Also their
types should be the same as declared in the prototype. For example if you put a = 4.5 while it
stands for int Length, the compiler would chop off the fractional part and will take it as 4 which
can lead to errors. Therefore, it is again reminded that data arguments be given in the same
order and type as given in the declaration of function prototype. If this rule is not followed
the program will give erroneous results.

RETURN STATEMENT
When we call a function the program control goes to the first line of the called function and
various statements of the function are evaluated. In case of void functions, the function ends at
the occurrence of the last closing right brace (}). Thereafter, the control automatically comes back
to calling function. It is optional to put the last statement in the function as
return;
In case of return type functions the return value is transferred to calling function through a
return statement and then at close of last right brace ( }) control passes on to calling function.
Therefore, the last statement in the return type function is as below.
return value;
For instance, in the definition of Weight the last statement is as below (in fact body has only
one statement).
 Functions ❖ 149❖
{return Length*Width*Height*Density;}
On encountering the last closing brace ‘}’ the control comes back to the main program at
the point where it left at the time of function call.
A function returns only one value. However by using pointers one may get to parametric
values and thus more than one values may be obtained (see Chapter 9 on pointers).
In Program 7.1 given below we illustrate a user defined function. It calculates the weight of
a cubical body. In this program the function Weight has been defined before the main () function
and hence, there is no need of function prototype.

PROGRAM 7.1 – Illustrates a user defined function which calculates the weight of a
prismatic bar.

#include<iostream>
using namespace std;
// Function defined above main
double Weight ( int L, int W, int H, float Rho)
{return L*W*H*Rho ; } //Rho is density of material
int main()
{
int A =2,B = 3,C = 1 ;
float D = 7.5;
cout<<“ Length =”<<A<< “, Width = ”<<B<< “, Height = ” <<C <<“, Density =
”<< D <<endl;
cout << “ Weight of bar = ”<<Weight(A,B,C,D)<< “\n ”;
// A, B, C and D are taken in the order L,W,H,Rho declared in function head.
return 0 ;
}

The expected output of the program is given below.

Length =2, Width = 3, Height = 1, Density = 7.5
Weight of bar = 45
In the following program, the function has been defined at the end of main() function. In
such cases it is required that function head followed by semicolon or prototype be declared before
the main().

PROGRAM 7.2 – Illustrates declaration of function above main and definition below the
main function.

#include<iostream>
using namespace std;
int Function (int x); //Function head or prototype in which
// it is not necessary to put the names of variable.
❖ 150 ❖ Programming with C++
int main()
{ int n;

cout << “ Number \t Square” <<endl;

for ( n = 0;n <=5;++n)
cout<< “ “<< n <<“ \t” << Function(n) <<endl;
return 0;
}

int Function(int x) // Function head

{ int Square = x*x; //Function body
return Square; }

The expected output is given below.

Number Square
0 0
1 1
2 4
3 9
4 16
5 25
In the following program function ptototype has been declared before main and function
definition is given after the main (). In the function prototype only the type of each parameter
separated by comma have been mentioned. The names of parameters need not be given in a
function prototype. A function prototype ends with a semicolon. The following program
calculates the sum of squares of two integers.

PROGRAM 7.3 – Illustrates placing function prototype before main() and definition at
the end of main()

#include<iostream>
using namespace std;

int Funct(int , int ) ; // Function prototype

// only types of parameters given.
int main()
{ int n , m;
cout << “n m Sum of squares” <<endl;

for (int i = 0;i<=5;i++)

{ n = i;
m = 2*i;
cout<< n <<“\t”<<m <<“\t”<< Funct(n,m) <<endl;}
 Functions ❖ 151❖
// Function called as Funct(n,m)
return 0;
}
// function definition
int Funct(int x, int y) // types and names both needed
{ int S = x*x + y*y; ; // in head
return S ; }

The expected output of the program is given below.

n m Sum of squares
0 0 0
1 2 5
2 4 20
3 6 45
4 8 80
5 10 125

In all the cases mentioned above you must have noticed that a function is not defined in-side
the main() function. Inside main() it is only called by mentioning the function name with
arguments in parentheses. For example, in the above program it is called as Funct(n,m);.
In the following program a function is defined to calculate the 4th power of an integer. Once
defined the function may be called any number of times.
PROGRAM 7.4 – Illustrates function definition above the main ().

#include <iostream>
using namespace std;
long power(int x) // function definition before main ()
{return x*x*x*x;}
int main()
{ int y = 2;
double z;
z = (power(y+2) –power(y))/ power(y);
cout <<“power(y) = ”<< power(y)<<“\n”;
cout<<“power(y+2) = ”<<power(y+2)<< “, and z = ” <<z<<endl;
return 0;
}

The program would give following result.

power(y) = 16
power(y+2) = 256, and z = 15
❖ 152 ❖ Programming with C++

7.4 INTEGRATION OF A FUNCTION

For integration of a function f(x) between the limits a and b, we follow the well known
trapezoidal rule as illustrated below.
b 1 1 nh 3
∫a f (x ) dx = h ( f +
2 0 1
f + f 2 + f 3 + … + f n–1 +
2
fn ) –
12
f ′′(ξ )
Where
h = (b – a )/ n
f0 = f (a)
fn = f (b)
fk = f (a + k × h)

The value ξ is a value between a and b. The last term goes to zero as n goes to ∞ . The method
is illustrated in the following program.
PROGRAM 7.5 – Illustrates integration of x2
#include <iostream>
#include <cmath>

using namespace std;

double integral(double x ,double a, double b )
{
int n = 100;

double F, h, Sum ;

h = (b–a)/n;
Sum = h *( b*b + a*a)/2;

for ( int k =1; k<n ; k++)

{x = a + k*h;
F = x*x;
Sum = Sum+ h *F ;}

return Sum ;
}

int main()
{
double y ;
cout<<“Integral= “<< integral(y,2.0,4.0) <<endl;
return 0 ;
}

4 1
The output is given below. It can be easily verified. ∫ 2
x 2 = (4 3 − 2 3 ) = 18.666 6666
3
Integral= 18.6668
 Functions ❖ 153❖

7.5 FUNCTIONS WITH EMPTY PARAMETER LIST

Up till now we used the function as formula type, in which some numerical value is calculated
and returned. But the function may, as mentioned in the introduction, just carry a message or
it may be used to print or display, etc. Such functions are generally of type void and may not
have any argument to act upon. Thus such functions may be defined as illustrated below.
void Display (void); // Display is the name of function
{cout << “You are learning C++.”<< endl;}
Or simply
void Display ();
{cout << “You are learning C++.”<< endl;}
In the following program a function is defined to display a message.

PROGRAM 7.6 – Illustrates a function with empty parameter list.

#include <iostream>
using namespace std;
void Display(void); //Function prototype, empty parameter list
void main()
{
Display(); // Program calls the function
}
void Display() //Function definition
{cout<< “Come! Let us learn C++.”<<endl;
cout<<“Have you any prior knowledge of C++?”<<endl;
cout<< “Or would you like to start from the beginning?” <<endl;
}

The expected output is given below

Come! Let us learn C++.
Have you any prior knowledge of C++?
Or would you like to start from the beginning?

7.6 FUNCTION OVERLOADING

The term refers to more than one function having the same name in a program. The overloaded
functions must have either different number of arguments or have arguments of different type.
Even if one argument is of different type in the overloaded functions the compiler will not give
an error signal.
This is illustrated in the following program in which three different functions having same
name Product () are declared. The compiler can find the appropriate function by comparing the
❖ 154 ❖ Programming with C++

arguments. The prototype of one function is int Product (int , int , int); , of the second is int
Product (int, int); which has different number of arguments though all are of type int and of
the third function is double Product ( int, double); . This function has same name but one
parameter is int while other is double, besides its type is also different, so it is different from the
other two.

PROGRAM 7.7 – Illustrates overloaded functions.

#include<iostream>
using namespace std;
int Product(int, int); //Three functions with same name
int Product (int,int,int);
double Product( int, double);
int main()
{int a = 5,b = 6, c= 2;
double d = 5.5;
cout << “Product of a and b = “ << Product(a,b) << endl;
cout<< “Product of a, b and c = “ << Product(a,b,c) <<endl;
cout << “Product of b and d = “ << Product(b,d)<< endl;
return 0;
}
int Product(int x, int y) // definition of function
{return x*y;}
double Product(int A, double B) // definition of function
{ return A*B;}
int Product(int n, int m, int k) // definition of function
{return n*m*k;}

The expected output is given below.

Product of a and b = 30
Product of a, b and c = 60
Product of b and d = 33

7.7 THE inline FUNCTIONS

When a program consists of a large numbers of functions considerable time of computer is wasted
in the function call routine that it has to follow. When a function is called, the processor saves
the present status of the current program, evaluates the function and returns to resume the
ongoing program. In case of frequently occurring small functions the overburden of function
call may be considerable and can make the program inefficient. At the same time writing of the
function code whereever it occurs in the program would make the program long and crude. It
is not a welcome feature even for maintenance of the software. So the concept of inline function
 Functions ❖ 155❖

is evolved. We define the function only once in the program. It is preceded with key word inline
as illustrated below. The compiler substitutes the code of the function whereever it is called in
the program. An illustration of an inline function definition is given below.
inline int Cube(x)
{return x*x*x ;}
The following program provides an illustration. In the program a general power function
has been defined for integers.

PROGRAM 7.8 – Illustrates inline function.

#include<iostream>
using namespace std;
inline int Power(int x, int n) // inline function
{ int P = 1; // definition
for (int i =0; i<n ; i++)
P *=x;
return P; }
int main()
{ int A = 2, B = 3, C =5;
cout<< “A square = ”<< Power(A,2); // function call
cout << “\nB to the power 4 = ”<< Power(B,4); // function call
cout<<“\nC to the power 3 = ”<< Power(C,3)<<endl;//function call
return 0;
}

The expected output is given below.

A square = 4
B to the power 4 = 81
C to the power 3 = 125

7.8 USE OF #define FOR MACROS

The pre-processor directive #define is generally used to create symbolic constants. It may also
be used to define very small functions or macros. Program 7.9 defines a macro for finding greater
of the two numbers. Program 7.10 given on the next page finds the area of a rectangle by defining
a macro. You will notice that no type is specified in the function.

PROGRAM 7.9 – A macro for finding maximum of two numbers.

#include <iostream>
using namespace std;
# define max(x,y) (x>y? x:y)
❖ 156 ❖ Programming with C++
int main()
{ int A, B;
cout<< “Write two integers :”; cin>>A>>B;
cout << “Maximum of the two = ” <<max (A,B)<< endl;
return 0;
}

The expected output is given below.

Write two integers :60 75
Maximum of the two = 75
The following program is another example of macro.

PROGRAM 7.10 – A macro for finding area of rectangle.

#include <iostream>
using namespace std;
#define Area_Rect(x,y) ((x)*(y))
int main()
{
double A, B;
cout<< “Write sides of rectangle :”; cin>>A>>B;

cout << “Area_Rect(A,B) = “ <<Area_Rect(A,B) << endl;

return 0;
}

The expected output is given below.

Write sides of rectangle :30.5 4.5
Area_Rect(A,B) = 137.25

FACTORIAL OF A NUMBER
Factorial of a number is also discussed in Section 7.11 of this chapter under the heading recursive
functions. The factorial n is also written as n!. It is defined below.
n! = n(n–1)(n–2)(n–3) …1
with 1! = 1 and 0! = 1
This can be achieved as below,
{int Factorial =1;
for ( int i =1; i<=n; i++)
Factorial *= i;
return Factorial; }
The factorial of a number may also be obtained as carried out in the following program.
 Functions ❖ 157❖
PROGRAM 7.11 – Illustrates function to evaluate factorial of a number.

#include <iostream>
long fact(int i)

{ if (i < 0) return 0;
int k = 1;
while (i > 0)
k *= i– –;
return k;
}
using namespace std;
int main()
{
int n,m;
long p;
m = 5;
n=3;
p = fact(m)/fact(m–n);
cout <<“Factorial 6 = ” <<fact(6)<<“, Factorial 4 = ”<<fact(4)<< “, p = ”
<<p<<endl;
return 0;
}

The output obtained by running the program is given below.

Factorial 6 = 720, Factorial 4 = 24, p = 60

7.9 C ++ STANDARD LIBRARY FUNCTIONS

The Standard C++ Library contains a large collection of predefined functions which may be
called and used by programmers. Here we take the functions belonging to the header file
<cmath>. However, there are many other functions in various header files in C++ standard library.
We have already got the experience of using the functions rand(),srand(),sqrt()and
swap() in the previous chapters.
For using a predefined C++ Standard Library function we have to include in our program
the header file in which the function is defined. Some of the predefined functions in <cmath>
header file are given in Table 7.1 below. In this table the arguments for the trigonometric
functions such as sin(x) or cos(x), etc., are in radians, i.e. the values of x is in radians. Similarly
the return value of function such as acos(y) = cos–1 (y) is also in radians.
There are a large number of other functions belonging to various header files such as
<iostream>, <string> , <set> ,<map>, etc. These are dealt in the relevant chapters.
❖ 158 ❖ Programming with C++

The following program illustrates, how to call C++ Standard Library functions and use them.
If the result is expected in floating point number, better take it as double.

PROGRAM 7.12 – The program illustrates evaluation of some functions of <cmath>.

#include <iostream>
#include <cmath>
using namespace std;

int main()
{

int a = 3600,b =10,c = –30 ,d = 3;

double A,B1,B2,B1f,B1c,Cabs ,Dexp;
A = sqrt(a); // returns square root of a number
Dexp= exp(d); // returns ed
B1 = log(b); // returns log of b to the base e
B2 =log10(b); // log(b) to the base 10
B1f=floor(B1); // returns lower integral value
B1c=ceil(B1); // returns just higher integral value.
Cabs =fabs(c); // returns absolute value

cout<< “ Value of e to the power3= Dexp = “ <<Dexp<<endl;

cout << “ log of 10 to base e = B1 = “ <<B1 <<endl;

cout << “ log of 10 to the base 10 = “ << B2 << endl;

cout << “ Floor value of B1 = “ << B1f << endl;

cout<< “ Ceil value of B1 = “<<B1c<<endl;

cout<<“ Absolute value of c = “<< Cabs <<“ and A = “<<A<< endl;

return 0;
}

The output of the program is as below.

Value of e to the power 3 = Dexp = 20.0855
Log of 10 to base e = B1 = 2.30259
log of 10 to the base 10 = 1
Floor value of B1 = 2
Ceil value of B1 = 3
Absolute value of c = 30 and A = 60
 Functions ❖ 159❖
The first line of output gives value of sqrt(a). The second line of output gives value of
e3. The third line of output gives log of 10 to the base e, the fourth line is the log of 10 to the
base 10 so value is 1. The floor value of B1 which is 2.30259 is the next lower integer and so
it is 2. Similarly the upper integer (ceil (B1)) value is 3. The value of c is – 30, so its absolute
value is 30.
Table 7.1 – Some functions in <cmath> header file

Function Description
Trigonometric Functions

cos(x) cosine of x, here x is in radians

sin(x) sine of x, here x is in radians
tan(x) tan of x, here x is in radians
acos(y) inverse of cosine, output in radians
asin(y) inverse of sine, output in radians,
atan (y) inverse of tangent, output in radians

Other Functions
round offs to higher integer value
ceil (x) Thus ceil (2.76 ) returns 3
exp(x) returns value of ex
fabs(x) returns absolute value of x
floor(x) rounds offs to lower integer value.
floor (2.76) returns 2
log(x) returns value of loge (x)
log10(x) returns value of log10(x)
sqrt(x) returns value of square root of x
pow( n , m) returns value of nm

The calling of trigonometric functions of C++ Standard Library is illustrated in the Program
7.13 below. The value of angle in all these functions is in radians.

PROGRAM 7.13 – The program illustrates evaluation of trigonometric functions.

#include <iostream>
#include <cmath>
using namespace std;
int main()
{
int a =60,b=45,c=30;
double alpha, beta, gama, C, S ,T ,theta, pi;
pi = 3.14159;
❖ 160 ❖ Programming with C++

alpha = a*pi/180; //converting angle from degrees to radians.

beta = b*pi/180;
gama = c*pi/180;

C = cos (alpha); // calling trigonometric functions

T = tan(beta);
S = sin(gama);

cout.precision(4); // number of digits desired in output

cout << “cos 60 = ” <<C <<endl;
cout << “tan 45 = ” << T << endl;
cout << “sin 30 = ” << S << endl;

theta = acos (0.5)*180/pi; // conversion of radians into degrees

cout << “cos inverse 0.5 = “ << theta << “ degrees” << endl;
return 0;
}

The expected output is given below.

cos 60 = 0.5
tan 45 = 1
sin 30 = 0.5
cos inverse 0.5 = 60 degrees
It may be reminded that for evaluations of trigonometric functions such as cosine, sine etc.
the angles are expressed in radians. In case of inverse functions such as acos() or asin(), etc., the
return values are also in radians. In the above program the angles are converted from degrees to
radians for evaluations of sine, cosine etc. and from radians to degrees in case of inverse
trigonometric functions.

7.10 PASSING ARGUMENTS BY VALUE AND BY REFERENCE

A reference to a variable is, in fact, an alias or another name for the same variable. Address of a
reference is the same as that of the variable to which it is a reference. A reference to a variable is
created by the declaration illustrated below.
int n = 10;
int& Number = n;

Here Number is reference to variable n. Similarly for other type of data we can create references
as given below.
double Price = 5.6 ;
double & Money = Price;
char ch = ‘B’.
char & kh = ch ;
 Functions ❖ 161❖

Here Money is a reference to Price and kh is a reference to ch. Dealing with a reference is as
good as dealing with the variable itself. If the value of reference is changed it also changes the
value of variable. For more details on the topic see Chapter 9 on pointers and references.
When we pass the arguments to a function by value we pass on the copies of values of the
variables to the function. The function may change the values of these copies but the values of
variables are not changed because the function does not know where the variables are stored and
hence cannot change values of variables. However, when we pass on a reference to a function,
any change in the value of reference changes the value of variable. Also note that address of a
variable, where its value is stored is same as that of its reference.
In the following program a function Swap() is declared and defined to interchange the values
of two variables. The values of the two variables x and y are passed on to the function. From
the output we see that though the function is alright but the values of x and y are not changed.
The values of their copies m and n are changed but x and y remain unchanged. In Program 7.15
given on the next page, the values are passed on to the function by references and we find the
values of x and y are now swapped.

PROGRAM 7.14 – Illustrates passing arguments by value to a function.

#include <iostream>
using namespace std;
inline void Swap ( int m, int n)
{
int Temp ;
Temp = m ;
m = n;
n = Temp;
cout<< “Address of m = ” << &m <<endl;
cout << “Address of n = ”<<&n <<endl;
cout<< “After swapping m = ” <<m <<“, n = ” <<n <<endl; }
/*The program illustrates User defined Swap function.It interchanges the
values.*/
int main()
{
int x ,y;
cout<<“Write two integers ”<<endl ;
cout << “x = ”;cin>>x; cout<<“ y = ”; cin>>y;
Swap( x, y);
cout <<“\nx = ”<<x <<“, y = ” << y << endl;
cout << “Address of x = ” << &x <<“, Address of y = ”<< &y <<endl;
return 0;
}
❖ 162 ❖ Programming with C++

From the output of the program, given below it is clear that values of m and n have been
swapped but the value of the variables x and y have not been changed. Also the addresses of m
and n are different from those of x and y.
Write two integers
x = 36
y = 66
Address of m = 0012FF24
Address of n = 0012FF28
After swapping m = 66, n = 36
x = 36, y = 66
Address of x = 0012FF7C, Address of y = 0012FF78
In the following program the values are passed of by references.

PROGRAM 7.15 – Illustrates passing arguments by references to a function.

#include <iostream>
using namespace std;
inline void Swap ( int& m, int& n)
//&m and &n are references to variables
{ int Temp ;
Temp= m ;
m = n;
n = Temp;
cout<< “Address of m = ” << &m <<endl;
cout << “Address of n = ”<<&n <<endl;
cout<< “After swapping m = ” <<m <<“, n = ” <<n <<endl; }
int main()
{ /*The program illustrates user defined Swap function. It exchanges the
values of variables.*/
int x ,y;
cout<<“Write two integers ”<<endl ;
cout << “x = ”;cin>>x; cout<<“ y = ”; cin>>y;
Swap( x, y);

cout <<“\nx = “<<x <<“, y = “ << y << endl;

cout << “Address of x = ” << &x <<“, Address of y = ”<< &y <<endl;
return 0;
}
 Functions ❖ 163❖

The expected output of the program is given below.

Write two integers
x = 36
y = 66
Address of m = 0012FF7C
Address of n = 0012FF78
After swapping m = 66, n = 36
x = 66, y = 36
Address of x = 0012FF7C, Address of y = 0012FF78

From the output note the following points.

(i) The values of the variables are now interchanged. Here &m and &n in the function Swap
( int & m, int & n) are references to x and y respectively.
(ii) The address of a reference is same as that of variable. For instance, the address of x is same
as that of m, similarly the address of y is same as that of n.

7.11 RECURSIVE FUNCTIONS

Recursion is a programming technique in which the function calls itself again and again till the
desired result is obtained. Naturally, in order to limit the number of times a function calls itself
there has to be parameter which increases or decreases every time the function is called and a limit
is put on its final value. The action is just like a loop. A recursive function is one that calls itself
directly or indirectly in its definition. For instance, let us consider factorial of a number n.
n ! = n (n–1)(n–2)(n–3) … 1
This can be put as
n! = n × !( n–1)
or Factorial n = n × Factorial (n–1)
In the above expression for the evaluation of factorial n, the function Factorial (n) calls itself
in the form of Factorial (n–1). If programmed in this manner it would be called recursive
function. In Program 7.11 we have already evaluated factorial n but in a non-recursive way.

PROGRAM 7.16 – Illustrates recursive function for determining a factorial.

#include<iostream>
using namespace std;
int F( int n )
{ if (n < 0) return 0;
if (n<=1) return 1;
else
return n* F(n–1);}
int main ()
❖ 164 ❖ Programming with C++
{
cout << “Factorial(6) = “ <<F(6)<<endl;
cout <<“Factorial (5) = “<<F(5)<<endl;
return 0; }

The expected output is given below.

Factorial(6) = 720
Factorial(5) = 120

FIBONACCI NUMBERS
Fibonacci numbers provide another example of recursive function. The numbers are in a series
as 0 1 1 2 3 5 8 13 21…
which can be expressed in the following way.
fibonacci(0) = 0
fibonacci(1) = 1
fibonacci(2) = 1
fibonacci(3) = fibonacci(3–1) + fibonacci(3–2) = 2
fibonacci(4) = fibonacci(4–1) + fibonacci(4–2) = 3
fibonacci(5) = fibonacci(5–1) + fibonacci(5–2)= 5
fibonacci(n) = fibonacci(n–1) + fibonacci(n–2)
The above sequence of numbers may be programmed as a recursive function as illustrated
in the following program.
PROGRAM 7.17 – Illustrates generation of Fibonacci numbers.
#include<iostream>
using namespace std;

int Fib(int n){ if (n ==0 || n <2) return n ;

else
return Fib(n–1) + Fib(n–2); }

main()
{
for(int N=0; N<=11; N++)
cout<< Fib(N)<<“ “;
return 0;
}

The expected output is given below.

0 1 1 2 3 5 8 13 21 34 55 89
The following program does calculation of cubes of a series of numbers as a recursive
function.
 Functions ❖ 165❖
PROGRAM 7.18 – Illustrates a recursive function for finding cubes of integers.

#include <iostream>
using namespace std;
void Recf(int x)
{ int cube ; //Definition of function
cube= x*x*x;
cout<< cube <<“ ”;
if (x> 5) exit(0);
Recf (x+1);}
int main ()
{
int x =1;
Recf (x);
return 0;
}

The expected output is given below.

1 8 27 64 125 216

EXERCISES
1. How is a function declared and defined?
2. Where is the user defined function placed in a program?
3. What is prototype of a function?
4. Explain the terms ‘passing arguments by reference’ and ‘passing arguments by value’.
5. Define a function to calculate the average value of a sample of values. Write a program to read
a sample of ten numbers, and display the average value by using a function.
6. How do you call a function?
7. Can you define a function inside main()?
8. How can you modularise a program with the help of functions?
9. Define a function that returns the area of a regular polygon. Make a test program to evaluate
the areas of an equilateral triangle and a square each having side equal to 10 units.
10. Define a function that accepts an integer as its argument and tests whether it is even or odd.
Make a program to test the function.
11. Make a program to integrate x n and test the program by evaluating integral of x3 between the
limits x = 5 and x =10.
12. The distance D between two points with co-ordinates ( x1, y1) and (x 2, y 2) on a plane is given
by
D = [( x2 – x1 )2 + (y2 – y1)2]1/2
❖ 166 ❖ Programming with C++

Define a function that returns D. Make a test program which uses the above function to
determine the distance between a point P(4, 8) and Q (7,12) and displays the same on the
monitor.

13. The distance D between two points in a three dimensional space is given by

D = [( x2 – x1 )2 + (y2 – y1)2 + (z2 – z1)2 ]1/2

Here x1, y1 , z1 and x2 , y2, z2 are the co-ordinates of the two points. Declare and define the
function for D. Also make program which uses the function by calling it from main(). Determine
and display the distance between points P( 5, 6, 8) and Q ( 10, 12, 30).
14. Write a program which displays the message “Come! Let us learn C++ together.”. Also write
a program to test your function.
15. Define a function with name sqrof() which accepts an integer and returns square of the integer.
16. Define a function which chooses the greatest of three integers. Make a program which reads
the three numbers from keyboard and displays the greatest number on the monitor.
17. Define a function with name powerof () which accepts two integers m and n as arguments and
returns n m.
18. Make a test program to test the function of Q.17 to determine the value of 44 and 232 .
19. Make a program that generates Fibonacci numbers and displays the first ten numbers.
20. Define a function to integrate sin(x). Make a test program to evaluate the integral between the
limits 0 to 180 degrees.
21. Make a program of a function which integrates a small polynomial between two specified limits.
22. What are macros? Give an illustrative example.
23. Find the mistakes in the following Swap function. Correct it and run it by making a test program.
inline void Swap ( int x, int y)
{ int temp ;
temp = x ;
x = y;
y = temp;}
❍❍❍
CHAPTER

8.1 DECLARATION OF AN ARRAY

An array is a sequence of objects of the same data type. In computer, array elements are stored
in a sequence of adjacent memory blocks. Each element of an array occupies one block of
memory, the size of which is allocated according to the data type as discussed in Chapter 3. The
declaration of one dimensional array is done as illustrated below.
type identifier [number of elements in array] ;
The above declaration of an array comprises from left to right the data type of elements of
the array, such as int, float, double, char, etc., followed by the name or identifier
of array, then the number of elements in the array put in square brackets. For example, an array
of 4 bills having values in whole numbers may be declared as below.
int Bill [4];
Here Bill is the name of array. It has 4 elements as indicated by 4 in square brackets. The
data type of its elements is int which is put in the beginning. But in this declaration we have
not given the values of elements, i.e. still the array has not been initialized. One method of
initializing the elements is by equating Bill [4] with values put in curly braces { } and separated
by comma as given below. Fig. 8.1 shows how a one-dimensional array is stored in computer
memory.
Bill [4] = {20, 30, 10, 15}; // initialization of an array
Index value of element
Array name Bill
0 1 2 3
Value of element 20 30 10 15
Memory of blocks

Fig. 8.1: One dimensional array in computer memory

The position of an element in array is called array index or subscript (Fig. 8.1). In case of
the array int Bill[4]; there are four elements. Their index or subscript values are 0, 1, 2
and 3. Note that count for array elements or subscripts starts from 0. Thus the first element of the
array is Bill [0], second elements is Bill [1], the third is Bill [2] and fourth is Bill [3]. When the
declaration and initialization of all the members are done together in same statement, we may
❖ 168 ❖ Programming with C++

or may not specify the number of elements in the square brackets. Thus the array Bill[4] may
also be declared and initialized as below.
int Bill[] = {20,30,10,15};
With the declaration of an array having n elements, the compiler will allocate n memory
blocks each of the size according to the data type for putting the values of elements of the array
when these are initialized. Figure (8.1) illustrates the storage of the members of Bill [4] in the
memory.
Arrays for other data types may similarly be declared and initialized. The array elements may
be of type int , float, double, char or it may be an array of pointers, objects of a class or classes.
For example an array with name Price consisting of 5 elements of type float, an array with name
Weight with 7 elements of type double and an array with name Address having 100 characters
may be declared as below.
float Price [5]; //The array can have 5 elements of type float
double Weight [7]; // The array can have 7 elements of type double
char Address[100]; // array can have 100 elements of type character

MULTI-DIMENSIONAL ARRAYS
If the elements of an array are arrays, the array is a multi-dimensional array. Let the 5 elements
of an array Names be arrays of 30 characters each. It makes ‘Names’ a two dimensional array
which may be declared as below.
char Names [5][30];
This array may be used to store 5 names each having up to 30 characters. Another instance
of a two dimensional array having elements of type int is illustrated below in Fig. 8.2. It is
declared and initialized as,
int M[2][5]= {5,2,3,2,4,6,7,8,9,8};
In computer memory a two dimensional array is also stored in the same way as shown in
Fig.8.1, however, for a physical picture Fig. 8.2 shows how a two dimensional array may be
visualised, i.e. two rows and 5 columns like a matrix.

M[2][5] 5 2 3 2 4

6 7 8 9 8

Fig. 8.2: A two dimensional array

Similarly a three dimensional array may be declared and initialized as below.
int S[2][3][4]={2,1,3,4,5,8,9,7,6,5,4,8,1,2,3,4,6,7,8,9,2,4,6,3}
S[0][3][4] S[1][3][4]

2 1 3 4 1 2 3 4

5 8 9 7 6 7 8 9

6 5 4 8 2 4 6 3

Fig. 8.3: A 3-dimensional array

 Arrays ❖ 169❖

For a physical picture we may visualise the array int S[2][3][4] as consisting of an array
of two elements each of which is a two dimensional array. This is illustrated in Fig. 8.3.

8.2 ACCESSING ELEMENTS OF AN ARRAY

An array can have elements of one type only. All the elements of an array may be either
integers, or floating point numbers or characters or class objects, etc., but they cannot be a
mixture of different types. An element of an array is a variable of the type declared in the array
declaration. It may be accessed by writing the name of array followed by index or subscript of
the element in square brackets. Thus we may call the first element of array Bill[4] as Bill[0], the
second element as Bill[1], the third element as Bill[2] and fourth element as Bill[3]. The ith
element of array Bill is written as Bill[i–1]; because i starts from 0. Similarly in case of two
dimensional array say M[2][5], we may call the an element belonging to ith row and jth column
as M[i–1][j–1], this is because both i and j start from 0.

PROGRAM 8.1 – Illustrates declaration and initialization of an array.

#include <iostream>
using namespace std;
main()
{ int Bill[4] = {20,30,10,15}; // Declaration and initialization
double Price[5] = { 5.6, 8.3, 7.0, 6.3,7.8 };
cout<<“Bill[0] = ”<< Bill[0]<<“, Bill[1] = ”<<Bill[1]<<endl;
cout<<“Bill[2] = ”<< Bill[2]<<“, Bill[3] = ”<<Bill[3]<<“\n\n”;

for ( int i =0; i<5;i++ ) // for loop for Price output

cout << Price [i] <<“ ”;

return 0;
}

The output of the program gives the values of individual elements of the array as follows.
Bill[0] = 20, Bill[1] = 30
Bill[2] = 10, Bill[3] = 15

5.6 8.3 7 6.3 7.8

If the number of values assigned are less than the total number of elements declared in the
array, zero is assigned to remaining elements. This is illustrated in the following program. If the
array Bill[4] is initialized as below,
Bill[4] = {20,30};
the first two elements will have values 20 and 30 respectively. The remaining two elements will
be assigned zero value.
❖ 170 ❖ Programming with C++

PROGRAM 8.2 – Illustrates the consequence of assigning less number of values than the
number of elements of array.

#include <iostream>
using namespace std;
int main()
{
int Bill[4] = {20,30};
// Two values are assigned while elements are four.
cout<<“Bill[0] = ”<< Bill[0]<<“, Bill[1] = ”<<Bill[1]<<endl;
cout<<“Bill[2] = ”<< Bill[2]<<“, Bill[3] = ”<<Bill[3]<<endl;
return 0;
}

The expected output of the above program is given below.

Bill[0] = 20, Bill[1] = 30
Bill[2] = 0, Bill[3] = 0
In the output you see that Bill[2] and Bill[3], i.e. third and fourth elements have been
assigned zero because in the initialization only two values are given. So the first two members
are assigned these values.
If in the initialization, the number of values are more than the number of elements of the
array, the compiler will show an error. See the following program.

PROGRAM 8.3 – Consequence of putting in more values than the array elements.

#include <iostream>
using namespace std;
int main()
{
int bill[4] = {20,30,12,21,4,32};
// six values written while elements are only four.
cout<<“ bill[0] = “<< bill[0]<<“, bill[1] = “<<bill[1]<<endl;
cout<<“ bill[2] = “<< bill[2]<<“, bill[3] = “<<bill[3]<<endl;
return 0; }

The expected output is the following error message.

too many initializers

8.3 INPUT/OUTPUT OF AN ARRAY

As described above, the input/output of an array is carried out element by element. Therefore,
either a for loop or while loop may be used for traversing the array. For example, if elements of
 Arrays ❖ 171❖

an array Bill[n] having n elements are to be displayed on the monitor, we may write the code as
below.
for (int i = 0; i<n; i++)
cout<< Bill[i]<<“ ” ;
The above statement will result in output of array elements in a single line, each element
separated by spaces as given in double quotes “ ”. If it is required to mention identifier for each
element along with its value, such as Bill[2] = 10 , we have to write the code as below.
for (int i =0; i<n; i++)
cout<< “Bill[“<<i<<” ] = “ <<Bill[i]<< “, ”;
An illustration of output of the above code for array Bill[4] of Program 8.1 is given below.
Bill[0] = 20, Bill[1] = 30, Bill[2] = 10, Bill[3] = 15,

USE OF CIN FOR AN ARRAY

For interactive input by user of the program we can use cin. Since each member of an array has
to be put in element by element so we may use a loop for the purpose. The following program
carries out input of 5 members of array bill [5].

PROGRAM 8.4 – Illustrates user interactive input of array elements.

#include <iostream>
using namespace std;
int main()

{
int bill[5];
cout<<“Enter five integers :”;
for (int i=0;i<5;i++)
cin>>bill[i];
cout<<“\n”;
for (int j=0;j<5;j++)
cout<<“ bill[”<<j<<“] = ”<< bill[j]<<endl;
return 0;
}

On clicking for running the program the following line will appear on the monitor with a
blinking cursor at the end.
Enter five integers :
At the place of blinking cursor type 5 integers each separated by a space and then press the
enter-key The following output will be displayed on monitor.
❖ 172 ❖ Programming with C++

Enter five integers :10 30 40 20 50

bill[0] = 10
bill[1] = 30
bill[2] = 40
bill[3] = 20
bill[4] = 50

In C++ there is no check to see that you do not overstep the number of elements in the array
while typing output statement. For example, if there are only 4 elements in the array you may
ask for fifth or sixth element, the compiler will not show an error but the output for 5th and
6th member would be some garbage value. It is illustrated below.

PROGRAM 8.5 – Illustrates output when the number of subscript values are more than the
number of array elements.

#include <iostream>
using namespace std;
main()
{
int bill[4] = {20,30,10,15};
cout<<“The bill amounts are as under.”<<endl;
for (int i =0; i <6; i++)
cout<<“bill[”<<i<<“] = ” << bill[i]<<“\n”;
return 0;
}

The output is given below.

The bill amounts are as under.
bill[0] = 20
bill[1] = 30
bill[2] = 10
bill[3] = 15
bill[4] = 1245120 // garbage value, because bill[] has only 4
bill[5] = 4333209 // elements. These values do not belong to array.

In order to make sure that only exact number of elements are entered in any function
accessing the array, the array size (number of elements) may be determined and used as illustrated
below for int bill[4].The code uses the function sizeof() which gives the number of byte
allocated. Let us take size = number of elements in array and it is determined as below.
 Arrays ❖ 173❖

size = number of bytes allocated for array/number of byte allocated to one element.
The code for determining size of an array with integer elements and using it in output
statement is illustrated below.
int size;
size = sizeof (bill) / sizeof(int);
for ( int i =0; i< size; i++)
cout << bill[i];
The application of this is illustrated in the following program.

PROGRAM 8.6 – Illustrates arithmetic operations on array elements.

#include <iostream>
using namespace std;
int main()
{double sum=0 , average =0;
double Price[5] ;
cout<< “Write the Quotations.”<<endl;
for ( int i =0; i<5;i++)
cin>> Price[i];
int size = sizeof(Price)/sizeof(double);
// size is the number of elements in array
cout<<“size = ”<<size<<endl;
for (int j =0; j<size ; j++) // use of size
sum += Price[j];
cout<<“Quotations are :”;
for(int k =0;k<size;k++) // use of size
cout<<Price[k]<<“ ”;

cout<<“\nThe number in the sample = ”<< size<<endl;

cout<<“sum = ” <<sum <<endl;

cout<< “Average price = ”<< sum/size <<endl;

return 0;
}

The expected output is given below.

Write the Quotations.
6.5 7.5 6.0 7.0 8.5
size = 5
Quotations are :6.5 7.5 6 7 8.5
The number in the sample = 5
sum = 35.5
Average price = 7.1
❖ 174 ❖ Programming with C++

DIFFERENCE BETWEEN AN ARRAY AND A C-STRING

The C-strings which are also arrays of characters are discussed in the next chapter, however, it is
worthwhile to bring out the differences between the two here. The following program brings
out the difference in their storage in computer. A string of characters declared and initialized as
char Name[] = { “Delhi”};
is an array of characters, however, the system appends a Null (‘\0’) character at the end of string
to mark the end of string. Therefore, the number of elements in the string increases by one, i.e.
in the memory for “Delhi” there are 6 elements – 5 of Delhi and one ‘\0’.
PROGRAM 8.7 – Illustrates difference between an array and C-string.
#include <iostream>
using namespace std;
main()
{
double Weight [] = {5.4, 6.35,12.6}; // An array
char Name[] = { ‘R’,’A’,’D’,’H’,’A’}; // An array
char ch[] = {“Radha”}; // A string. It can also be coded as
// char ch [] = “Radha”;
cout<<“Size of Name = ” << sizeof(Name)<<endl;;
cout << “Size of ch = ” << sizeof (ch)<<endl;
for (int i=0;i<5;i++)
cout <<Name[i] ;
cout<<“\n”;
for (int j=0;j<3;j++)
cout<<Weight[j]<<“ ”;
cout<<“\n”;
cout <<ch ;
return 0;
}

The expected output is as below.

Size of Name = 5
Size of ch = 6
RADHA
5.4 6.35 12.6
Radha
The above program shows that a char array may be initialized by individual characters as well.
In that case it is an array of characters without the Null character at the end. However, if a string
of characters is assigned in any one of the following two methods the compiler appends the Null
character at the end of string of characters to mark the end of string.
char ch[] = {“Radha”}; // or as below
char ch[] = “Radha”;
 Arrays ❖ 175❖

That explains that the difference between the array char Name[] and char ch [] is the Null
character that is appended by the system at the end of string. Thus number of elements of string
in the case of char ch[] = “Radha”; is one more than that in Name[]. This is because when we
put it as a string in double quotes as in “Radha” the compiler attaches the Null (‘ \0’ ) character
at the end of the string, to mark the end of string. Moreover a string is treated as a single unit.

SWAPPING ELEMENTS OF TWO ARRAYS

As already mentioned the array as a whole cannot be assigned. Therefore, arrays cannot be
swapped as a whole with another array. However, the swap operation may be carried out by
swapping element by element. This is illustrated below.

PROGRAM 8.8 – Illustrates swapping of elements of two arrays.

#include <iostream>
using namespace std;
int main()
{
int bill[4] = {20,30,10,15};
int Bill[5] = { 5, 6 ,7, 8,9};
for ( int i =0; i<5;i++)
{swap ( bill[i] , Bill[i] );
cout <<“ bill[”<<i <<“] = ”<< bill [i] <<“, ” ; }
cout <<“\n”;
for (int j =0; j<5; j++)
cout << “Bill[”<<j<<“] = ”<< Bill[j]<<“, ” ;
return 0;
}

The expected output of the program is given below.

bill[0] = 5, bill[1] = 6, bill[2] = 7, bill[3] = 8, bill[4] = 9
Bill[0] = 20, Bill[1] = 30, Bill[2] = 10, Bill[3] = 15, Bill[4] = 1245120

The Bill[4] is a garbage value. For swapping elements of arrays, take care that arrays are of
equal size otherwise such potentially dangerous errors may happen.

INTERACTIVE INPUT FOR SEVERAL ARRAYS

The elements of more than one array may also be entered by user of the program in an interactive
program. As illustrated above we use cin for the purpose. A for loop or while loop may be
used for each array. For example let us name an array as data [50 ]. For interactive input the
following code may be used.
for (int i=0;i<50;i++)
cin>>data[i];
❖ 176 ❖ Programming with C++

This has been already illustrated in the Program 8.6 wherein different types of data have been
entered interactively by user. Also more than one array may be declared in same line if their
elements are of same type.

PROGRAM 8.9 – Illustrates user interactive input of several arrays.

#include <iostream>
using namespace std;
int main()
{
int Bill[4], bill[5]; // both arrays are of type int
char Myname[7];
cout<<“Write values of array Bill ; ”;
for ( int i = 0; i<4; i++)
cin >> Bill[i];
cout<<“Write values of array bill ; ”;
for ( int k = 0; k<5; k++)
cin >> bill[k];
cout<<“Write values of array Myname ; ”;
for ( int n = 0; n<7; n++)
cin >> Myname[n];
cout<<“The Bill elements are: ”;
for (int j=0;j<4;j++)
cout<< Bill[j]<<“ ”;
cout<<“\n” ;
cout << “The bill elements are ”;
for (int m=0;m<5;m++)
cout <<bill[m]<<“ ” ;
cout<<“\n ”;
cout<< “Myname is : ”;
for (int p=0;p<7;p++)
cout <<Myname[p];
cout<<“\n”;
return 0; }

While running the above program, the following line will first appear on the monitor.
Write values of array Bill ;
Type four integers each separated by a space from the previous one, say, 10 20 40 30 are typed.
Next press enter-key. The second line as given below will then appear on the monitor.
Write values of array bill ;
 Arrays ❖ 177❖

As done before for array Bill, type five integers each separated by space and press enter-key.
Then the third line given below will appear.
Write values of array Myname ;
Type a name consisting of 7 characters and each letter separated by a space. Then press enter-
key. The following output will appear on monitor. In this the first three lines are of input and
last three lines are of output.
Write values of array Bill ; 10 20 40 30
Write values of array bill ; 15 30 50 40 60
Write values of array Myname ; N a t a s h a
The Bill elements are: 10 20 40 30
The bill elements are 15 30 50 40 60
Myname is : Natasha

8.4 SEARCHING A VALUE IN AN ARRAY

Often it is required to search a value in an array. This may be carried out by comparing the key
value with each element of array. It is illustrated in the program given below.

PROGRAM 8.10 – Illustrates searching a value in an array.

#include <iostream>
using namespace std;
int main()
{
int sum=0;
int bill[ ] = {20,30,10,15, 50,40 ,30, 70,95};

int x ;
int Size = sizeof (bill)/sizeof(int);
cout<<“Write the number to be searched:”;
cin >>x; // x is the value to be searched in the array

for(int i =0; i<Size;i++)

if( x == bill[i])
{cout<< “Yes, the number is in bill- bill[“<<i <<“]= “<<bill[i]<<endl;
goto End;}
cout <<“The number is not in bill” <<endl;
End:
return 0;
}

The output is as under. First we enter the number 50 and the output is below.
Write the number to be searched:50
❖ 178 ❖ Programming with C++
Yes, the number is in bill- bill[4]= 50
In the second trial we enter the number 75 and the output is as below.
Write the number to be searched:75
The number is not in bill

BINARY SEARCH FOR ORDERED ARRAYS

In the above search method every array element is compared with the key. In case of ordered
arrays the search may be shortened considerably by binary search which applies only to sorted
arrays. In this method first determine the number of elements in the array. Then determine in
which half of the array the key value (the value we wish to search) lies. This is easily done by
comparing with the element at middle because the array is an ordered array. Then search only
that half in which the value lies. This half may again be divided into two halves and again it is
determined in which half ( ¼ the of original array) the key lies. In this way the search continues
till the value is found. For large ordered arrays this method is very efficient. The following code
illustrates the method. If the value is not in the array the following code displays it in the output
Also C++ Standard Library has an algorithm binary_search() (see Chapter 25) and a programmer
may use it by including the header file <algorithm> in the program.
PROGRAM 8.11 – Binary search of an array for a value.

#include <iostream>
using namespace std;
int main()
{ int sum=0;
int bill[ ] = {20,30,40, 50,60 ,70, 80,95, 102, 166, 175,200 };
int x ;
cout<<“Write the number you want to find :”; cin>>x;
int Size = sizeof (bill) / sizeof(int);
int Low =0;
int High = Size -1 ;
while ( Low <= High)
{
if(Low ==High && bill[High] != x)
{ cout<<“The number is not in the array.\n”;
break;}
int Mid = (Low + High)/2;
if ( bill[Mid] == x)
{cout<< “Value found. It is bill [”<<Mid <<“]”<< endl;
break;}
else
if ( bill[Mid] > x )
High = Mid–1;
else
Low = Mid+1;}
return 0;
}
 Arrays ❖ 179❖

The first trial gives the following output.

Write the number you want to find :70
Value found. It is bill [5]
The second trial with element not in array gives the following.
Write the number you want to find :103
The number is not in the array.

8.5 ADDRESS OF AN ARRAY

With the declaration of array as int Bill[4]; four blocks of memory are allocated by
the compiler for the array elements, i.e. one block for each element for storing its value. The array
is initialized as given below. Figure 8.4 shows the memory blocks and the values expressed in
binary in each block.
int Bill[4] = {20,30,10,15};
The size of blocks depends on the type of data in the array. For the present case, the type
of elements is int so 4 bytes are allocated for each block. The value of each element is stored in
binary as illustrated in Fig. 8.4 below.
Array element 0 1 2 3

10100 11110 1010 1111

Memory address 0012FF 70 0012FF 74 0012FF 78 0012FF 7C

Fig. 8.4: An illustration of array address and addresses of array elements.
The values inside the blocks are values of array elements in binary

Here 10100, 11110, 1010 and 1111 are binary equivalents of 20, 30, 10 and 15 respectively,
and the memory addresses written below the blocks are in hexadecimal.
The address of an array is the address of first element of the array which is the number of
the first byte of the memory block allocated to the first element of the array. It may be extracted
by simply calling the name of array as illustrated in the following code. Thus for array Bill[4]
the code for getting address of the array is
cout << Bill;
The output will be the address of the array. The address of any element of an array may also
be extracted by calling array (name + offset). The offset is equal to the subscript or index value
of the element. Thus for getting the address of the second element of array Bill[4] we can call
Bill+1 as illustrated below.
cout << Bill+1 ;
Similarly, for obtaining the address of third element code is as below.
cout << Bill+2;
One can write similar codes for other elements as well.
The following program determines the address of an array and addresses of individual
elements of the array.
❖ 180 ❖ Programming with C++

PROGRAM 8.12 – Addresses of an array and its elements.

#include <iostream>
using namespace std;
int main()
{ int Bill[4] = {20,30,10,15};
cout<< “The address of the array is ”;
cout<< Bill <<“\n”;
cout<< “The addresses of the four elements are as below.”<<endl;
cout<<“ ”<<Bill <<“ , ”<<Bill+1 <<“ , ”<<Bill+2<<
“ , ”<<Bill+3<<endl;
return 0;
}

The expected output of the above program is as under.

The address of the array is 0012FF70
The addresses of the four elements are as below.
0012FF70, 0012FF74, 0012FF78, 0012FF7C

The above output shows that elements of array Bill[4] are stored in computer memory
starting from byte number 0012FF70 (in hexadecimal ). Each element of array in this case occupies
four bytes. The address of an array is the address of the first element of the array. Therefore,
the address of array is 0012FF70. The subscript values of elements are 0, 1, 2 and 3. The addresses
of elements are 0012FF70, 0012FF74, 0012FF78 and 0012FF7C. These are the byte numbers
or addresses of the first bytes of the memory blocks allocated for each array element. Each
element, in this case, is allocated a memory block of 4 bytes.
In the above programs we have written the individual elements of the array Bill[4]as Bill[0],
Bill[1], etc. Also, we could as well write them as 0[Bill], 1[Bill], etc. This is illustrated in the
following program. As already mentioned above the numbers 0, 1, 2 and 3 are called subscripts.
You should note that subscripts start from 0. The starting element of array is the 0th element,
that’s why the array of 4 elements has the subscripts 0, 1, 2 and 3. This is true for all arrays. The
programmers generally make mistake by taking the starting element as 1.
Also remember that a subscript value represents the position of an element in the array.
It is not the value of the array element.

8.6 ARITHMETIC OPERATIONS ON ARRAY ELEMENTS

Individual array element can be treated as any other variable of the same type. Therefore, all
the operators which can be applied on other variables of same type can also be applied to the
elements of the array. The arithmetic operators, i.e. +, –, *, / and % may be used on individual
array elements just like these operators are used on other variables. Program 8.13, given below
illustrates the arithmetic operations on array elements. The program determines the average value
and standard deviation of a sample of data presented as elements of an array. In quality control
 Arrays ❖ 181❖

we often take a sample of measurements of objects and find out the average value and the standard
deviation.

PROGRAM 8.13 – Illustrates arithmetic operations on array elements.

#include <iostream>
#include<cmath>
using namespace std;
int main()
{
int m,n,p ;
double Sum = 0 ,Average = 0 , Sigma = 0;
int Sample[] = {20,16,10,15,11,12,15,22,14,13,14};
m =sizeof(Sample[1]); // gives number of bytes for one element
p = sizeof(Sample);
//p is the total number of bytes used for array.
n = p/m;
// n = number of elements in the array.
for(int i=0;i<=(n–1);i++)
Sum += Sample[i]; // Calculates the sum of all elements.
Average = Sum/n;
double Var =0;
for ( int j =0; j<n;j++)
Var += ((Sample[j]–Average)*(Sample[j]–Average));
Sigma = sqrt(Var/(n–1));
// Here sigma is the standard deviation
cout<<“Sum of elements = ”<<Sum<<“\nNo. of elements = ”<<n <<endl;
cout <<“Sample average = ”<<Average <<endl;
cout<<“Var = ”<<Var <<endl;
cout<< “Sample std. deviation = ”<< Sigma<<endl;
return 0;
}

The expected output is given below.

Sum of elements = 162
No. of elements = 11
Sample average = 14.7273
Var = 130.182
Sample std. deviation = 4.65717
The following program illustrates the general arithmetic operations on array elements.
❖ 182 ❖ Programming with C++

PROGRAM 8.14 – Another illustration of arithmetic operations on array elements.

#include <iostream>
using namespace std;
int main()
{
float B[4] = {2.5 ,3.4 ,6.0,12.2};
float C[4] = {4.0,2.0,3.6,6.4};
float R[4];
double S[4],a,b,c,d, sum =0;
a = B[2] + C[3];
b =B[3]– C[1];
c= B[2] * C[1];
d = (B[3]+C[2]) /(C[1]*B[1]);
cout<< “a = ”<<a<<“,\tb = ”<<b <<“\n”;
cout<<“c = ”<<c<< “,\t\td = ”<<d<<“\n”;
for (int i=0;i<=3;i++)
{S[i] = B[i] /C[i];
R[i] = B[i] * C[i];
cout<< “S[”<<i<<“] = ”<< S[i]<< “,\tR[”<<i<<“] = ”<< R[i]<<endl; }

for (i=0;i<=3;i++)
sum += (B[i] +C[i]);
cout << “Sum of B and C array elements = ” <<sum <<endl;
return 0;
}

The expected output of above program is given below.

a = 12.4, b = 10.2
c = 12, d = 2.32353
S[0] = 0.625, R[0] = 10
S[1] = 1.7, R[1] = 6.8
S[2] = 1.66667, R[2] = 21.6
S[3] = 1.90625, R[3] = 78.08
Sum of B and C array elements = 40.1

OPERATION OF FUNCTIONS ON ARRAY ELEMENTS

In the following program two functions are defined, one returns the square of a number and
other returns square-root of the number. The functions operate on the array elements and return
the square and square root of the array elements.
 Arrays ❖ 183❖

PROGRAM 8.15 – Application of C++ Standard Library functions on array elements.

#include<iostream>
#include<cmath>
using namespace std;
int F1(int x)
{return pow(x,2);
}

double F2(int n)
{return sqrt(n);}

main()
{int A[4] = { 2,3,4,5};
int B[4];
double S[4];

for(int i=0;i<=3;i++)
{B[i] = F1(A[i]);
S[i] = F2(A[i]);

cout<<“A[”<<i<<“] = ”<<A[i]<< “\tB[”<<i<<“] = ”<<B[i]<<“,\tS[” <<i<<“] =

”<<S[i]<<“\n”;}
return 0;
}

The expected output is given below.

A[0] = 2 B[0] = 4 S[0] = 1.41421
A[1] = 3 B[1] = 9 S[1] = 1.73205
A[2] = 4 B[2] = 16 S[2] = 2
A[3] = 5 B[3] = 25 S[3] = 2.23607

8.7 SORTING OF ARRAYS

In many cases it is required to sort an array in the ascending or descending order. For example
you may want to arrange the marks obtained by students in a descending order so that the student
who scores highest is at the top and the one with lowest marks is at the bottom, or you may
like to arrange a list of names in an alphabetical order, etc. The Program 8.16 illustrates the code
for sorting out a list of numbers entered by user.
If the list is to be arranged in ascending order the element with lowest value should be the
first. Let kim [j] and kim [j+1] be the two adjacent members of the array kim[]. Now the two
❖ 184 ❖ Programming with C++

are compared and if kim [j] > kim [j+1] the two are swapped. The comparison then shifts to
next element and so on to the end of array. In this way the element with highest value goes to
the end.
But if kim [j] is less than kim [j + 1] then no swapping is done and comparison shifts to
next, i.e. kim[j + 1] and kim[j + 2]. This process of comparison and swapping is successively
carried out with all the elements of the array. Thus if n is the number of elements in the array
the process of comparison will be repeated n – 1 times. And this will have to be carried out
(n –1) times to complete the sorting. The following program is coded to illustrate the process
of sorting in an ascending order.

PROGRAM 8.16 – Illustrates sorting of an array of characters.

#include <iostream>
using namespace std;

int main()
{
const int p = 5;
char kim[10] ;

cout <<“Enter 5 characters: ”;

for(int i=0;i<p;i++)
cin>> kim[i];

cout<<“\nyou have entered the following characters:\n ”;

for(int k =0;k<p;k++)
cout<<“ ”<<kim[k]<<“ ”;

cout<<“\n”;
for ( int n = 0 ; n<p; n++)

{for(int j =0;j<p–1;j++) //for explanation of p-1 see output

{if(kim[j]>kim[j+1])
swap( kim[j],kim[j+1]); }

for ( int r =0 ; r<5; r++)

cout<< kim[r]<< “ ”;
cout << “\n”;}
return 0;
}

The expected output is as below. The first line of output depicts that five characters, i.e. E
D C B and A are entered. The next line shows that the first character E is successively compared
with the other characters and is placed at the end of array. The third line of output shows that
the new first element D is similarly compared and placed at a position before E. We see that in
 Arrays ❖ 185❖

4th round that is 1 less than the size of array all are sorted. Notice that two for loops have been
used in the program. One for comparing the values and placing the element with higher value
at appropriate place. The second for loop is for repeating the above process a number of times.
Enter 5 characters: E D C B A
you have entered the following characters:
EDCBA
D C B A E // E goes to the end in first round.
C B A D E // D goes to position one less than the end.
B A C D E // the C is placed in its position, i.e. on left of D.
A B C D E // B is placed after A , Total 4 cycles of sorting.
A B C D E

The following program illustrates the sorting of a list of numbers in ascending order.

PROGRAM 8.17 – Illustrates sorting of a list of integers.

#include <iostream>
using namespace std;
main()
{
int kim[10] ;
cout <<“Enter 10 integers: ”;
{ for(int i=0;i<10;i++)
cin>> kim[i]; }

cout<<“The sorted list is given below \n ”;

{for (int i=0;i<10;i++)
for(int j =0;j<9;j++)
if (kim[j]>kim[j+1])
swap(kim[j],kim[j+1]); }

for(int i=0;i<10;i++)
cout <<“ ”<< kim[i]<<“ ”;
cout<<“\n”;

return 0;
}

The output is given below.

Enter 10 integers: 23 11 43 56 78 65 43 22 88 37
The sorted list is given below
11 22 23 37 43 43 56 65 78 88
❖ 186 ❖ Programming with C++

8.8 FINDING THE MAXIMUM/MINIMUM VALUE IN AN ARRAY

Often it is required to find the maximum and minimum values in a list. For instance, in a list
of marks obtained by students it is often needed to find out the topper. The listing of the
following program determines the maximum value in an array of values of elements. It also finds
which element has the maximum value. The values of the array elements are entered by the user
of the program.
The scheme of the program is that let the first elements be assumed as the maximum value.
This is compared with the second element. If the second element is greater than the first, the
second element is designated as maximum and compared with the third and same process is
repeated. If the first element is greater than second, the comparison shifts to third without doing
anything. The process is carried out with all elements and maximum value is thus found.

PROGRAM 8.18 – Finding the array element with maximum value in an array.

#include <iostream>
using namespace std;
int main()
{int kim[10] ={0},max ;
cout <<“Enter 10 integers: ”;
{for(int i=0;i<10;i++)
cin>> kim[i];
}
cout<<“\nyou have entered the following numbers: \n”;
for(int i =0;i<10;i++)
cout<<“ ”<<kim[i]<<“ ”;
cout<<“\n ”;
max = kim[0];
{for(int j =0;j<9;j++)
if(kim[j]>max)
max=kim[j];
else max = max; }
cout <<“max = ”<< max<<endl;
{for(int k=0;k<10;k++)
if ( max ==kim[k])
cout<<“kim[”<<k<<“] = ” << kim[k] <<endl; }
return 0;
}

The expected output is as under.

Enter 10 integers: 65 34 23 87 68 97 64 32 10 45
you have entered the following numbers:
 Arrays ❖ 187❖
65 34 23 87 68 97 64 32 10 45
max = 97
kim[5] = 97
The output gives the maximum value and the index value of element. The index values start
from 0 in arrays, so 97 is 6th element with index value 5.

8.9 PASSING AN ARRAY TO A FUNCTION

Just like a variable, an array may also be a parameter of a function. For declaration of a function
with array as parameter we have to mention data type of the array and an unsigned integer which
stands for number of elements in the array. Please note that the array elements cannot be
functions. Thus the prototype of a function with an array as a parameter may be written as below.
int Func(int [], size_t ); // Prototype
In the above expression size_t is the typedef of unsigned int. And for the function definition,
one may write as below.
int Func(int A [], size_t n)
{statement;
————————
————————;}

PROGRAM 8.19 – Illustrates passing of an array to a function.

#include <iostream>
using namespace std;
double Average (double[], size_t ); // Prototype of the function
//Average
int main()
{int n=10;
double kim[10];
cout <<“Enter 10 floating point numbers: ”;
for(int i=0;i<10;i++)
cin>> kim[i]; // entering the array elements
cout <<“You have entered the following numbers”<<endl;
for(int j=0;j<10;j++)
cout<< kim[j]<<“ ”;
cout << “\n Average = ” << Average(kim,n)<< endl;
return 0;
}
double Average(double kim[], size_t n)
// definition of function
{ double Total = 0;
❖ 188 ❖ Programming with C++
for (int i =0; i<n; i++)
Total += kim[i];
return Total / n;
}

The expected output is given below.

Enter 10 floating point numbers: 6.5 4.6 5.8 5.9 6.1 6.7 6.4 5.3 4.6 5.0
You have entered the following numbers
6.5 4.6 5.8 5.9 6.1 6.7 6.4 5.3 4.6 5
Average = 5.69
In the following program two arrays are passed to a function. The function evaluates the
products of their respective elements and calculates the sum.

PROGRAM 8.20 – Illustrates passing of more than one array to a function.

#include <iostream>
using namespace std;
double Product (double [], double[], size_t, size_t);
//function prototype
int main()
{ int n =4, m=4;
double kim[4];
double Bill[4];
cout <<“Enter 4 decimal numbers for kim: ”;
for(int i=0;i<4;i++)
cin>> kim[i];
cout<<“\nEnter 4 numbers for Bill: ”;
for(int j=0;j<4;j++)
cin>> Bill[j]; // Entering array elements
cout <<“You have entered the following numbers for kim.”<<endl;
for(int k=0;k<4;k++)
cout<< kim[k]<<“ ”;
cout<<“\nYou have entered the following numbers for Bill ”<<endl;
for(int s=0;s<4;s++)
cout<< Bill[s]<<“ ”;
cout<<“\n Sum of Product of array elements = ” << Product(kim, Bill, n,
m)<< endl;
return 0;
}
double Product(double kim[],double Bill[], size_t n, size_t m )
{double Product[] ={0,0,0,0};
 Arrays ❖ 189❖
double Total=0;
for ( n =0; n<4; n++)
{Product[n]= kim[n] * Bill[n];
Total += Product[n]; }
return Total;
}

The expected output is given below.

Enter 4 decimal numbers for kim: 1 2 3 4
Enter 4 numbers for Bill: 5 6 7 8
You have entered the following numbers for kim.
1234
You have entered the following numbers for Bill
5678
Sum of Product of array elements = 70

8.10 TWO DIMENSIONAL ARRAYS

The elements of an array may be arrays. Such arrays are called multi-dimensional arrays. Thus
an array having elements which are one dimensional arrays, is a two dimensional array. Similarly,
an array whose elements are two dimension arrays is a three dimensional array. The declaration
and initialization of multi-dimensional arrays have already been illustrated in the beginning of
this chapter. A two dimensional array may be declared as given below.
int bill[2][4];
The array may be taken as having two rows of 4 columns each like in a matrix.
There are many applications of two dimension arrays in mathematic, in science, in
engineering and even in making list of names, etc. The following program illustrates the input
and output of a two dimensional array. For input or output nested for loops may be used. Say
for bill[2][4], for the outer loop the subscripts vary from 0 to 1 for the two rows and for the
inner loop the subscripts vary from 0 to 3 for the four columns. For each value of outer subscript
the inner one has values 0, 1, 2 and 3. See the following program.

PROGRAM 8.21 – Illustrates input/output of two dimensional arrays.

#include <iostream>
using namespace std;
void read(int bill[2][3]);
int main()
{
int bill[2][3];
read(bill);
❖ 190 ❖ Programming with C++
for(int i=0;i<2;i++)
{ cout << “Row ”<<i+1<<endl; ;
for(int j=0;j<3;j++)
cout<<“bill[”<<i<<“][”<<j<<“] = ”<< bill[i][j]<<“ ”;
cout<< “\n”;}
return 0;}
void read(int bill[2][3]) // definition of function read()
{cout<< “Enter 6 integers , 3 per row;\n”;
for(int k=0;k<=1;k++)
{cout<<“Row ”<<k+1<<“ ; ”;
for(int j=0;j<3;j++)
{
cin>>bill[k][j]; }}}

The expected output is given below.

Enter 6 integers , 3 per row;
Row 1 ; 11 12 13
Row 2 ; 21 22 23
Row 1
bill[0][0] = 11 bill[0][1] = 12 bill[0][2] = 13
Row 2
bill[1][0] = 21 bill[1][1] = 22 bill[1][2] = 23

The initialization of a two dimensional array may also be done as it is carried out for one
dimensional array illustrated in Program 8.1 or with cin as illustrated in the above program. The
elements in all types of arrays are stored in the computer memory in consecutive memory blocks.
Each element occupies one block of memory. The size of memory block depends on the type
of array elements. In the following program the address of each element of a two dimensional
array has been called. The output shows how a two dimensional array is stored in computer
memory.

PROGRAM 8.22 – Addresses of elements of multidimensional arrays.

#include <iostream>
using namespace std;
int main()
{
short bill[2][3];
cout << “Enter 6 small numbers : ” ;
/*you may write in any groups, the program will read left to right. The
numbers must be separated by white space and not by comma.*/
for(int i=0;i<2;i++)
{for(int j=0;j<3;j++)
 Arrays ❖ 191❖
cin>>bill[i][j];
}
cout<<“/n Address of the Array is “<<bill<<endl;
cout<<“\n ”;
cout<<“Addresses of members of 2 dimension array are :\n”<<endl;
{for(int i=0;i<2;i++)

{for(int j=0;j<3;j++)
cout<<“bill[”<<i<<“][”<<j<<“] = ”<< &bill[i][j] <<“\n”; }
}
return 0;
}

The expected output is viven below.

Enter 6 small numbers : 3 4 5 6 7 8
Address of the Array is 0012FF74

Addresses of members of 2 dimension array are :

bill[0][0] = 0012FF74
bill[0][1] = 0012FF76
bill[0][2] = 0012FF78
bill[1][0] = 0012FF7A
bill[1][1] = 0012FF7C
bill[1][2] = 0012FF7E

From the output it is clear that elements of first row are stored first in adjacent memory
blocks followed by elements of second row. The array may as well be written as
3 4 5
Bill [2][3] =
6 7 8

8.11 TWO DIMENSIONAL ARRAYS AND MATRICES

An important application of two dimensional arrays is in representation and operations on

matrices. For example, a matrix A = ajk with ‘j’ rows and ‘k’ columns may be represented by a
two dimensional array A [j][k].
ADDITION OF TWO MATRICES
Let A =( a ij) and B = (b ij) be the two matrices. The addition and subtraction of A and B are
illustrated below.
Addition
C=A+B
Or Cij = aij + bij
Subtraction
❖ 192 ❖ Programming with C++
D=A–B
Or Dij = aij - bij
In both addition and subtraction the corresponding components of the matrices are added
and subtracted. When presented in form of arrays the codes for addition/subtraction of matrices
having n rows and m columns may be written as follows.
for( int i = 0 ; i<n ; i++)
for(int j = 0; j<m; j++)
{C{i][j] = A[i][j] + B[i][j];
D[i][j] = A[i][j] – B [i][j];}

PROGRAM 8.23 – Illustrates addition/subtraction of matrices.

# include <iostream>
using namespace std;
int main()
{int C[2][3] , D[2][3];

int A[2][3] = {20,30,40,50,60,70};

int B [2][3] = {10,20,20,30,20,10};

for ( int i =0; i<2; i++)

for ( int j = 0; j<3; j++)
{C[i][j] = A[i][j] + B[i][j]; // addition
D[i][j] = A[i][j] – B[i][j];} //subtraction

for ( int n =0 ; n<2; n++)

{for ( int m = 0; m<3; m++)
cout<<C[n][m]<<“ ” ;
cout << “\n”;}

cout<<endl;

for ( int p =0 ; p<2; p++)

{for ( int s = 0; s<3; s++)
cout<<D[p][s]<<“ ” ;
cout << “\n”;}
return 0 ;
}

The expected output is given below.

30 50 60
80 80 80

10 10 20
20 40 60
 Arrays ❖ 193❖

MULTIPLICATION OF TWO MATRICES

Let A and B be the two matrices in which number of columns of A equals number of rows of
B, i.e.
A = ( aij)
B = (bj m)

C = AB
Or cim = aij . bj m
The repetition of subscript j indicates summation of terms for the range of values of j. For
sake of illustration let us take A be 2 × 3 matrix and B be as 3 × 1 matrix and let
 a11 a12 a13 
A= 
 a21 a22 a 
23
and
 b11
 
B =  b21
 b 
31
The product is a 2×1 matrix which is given as below
c
C =  11  = AB = (a2m) (bm1)
 c 
21

 b11
 c11   a11 a12 a13  a b + a12 b21 + a13 b31 
 b  =  11 11
 c  =  a a22 a23   21  a21 b11 + a22 b21 + a23 b31 
21 21
 b31

The following program illustrates the above product.

PROGRAM 8.24 – Illustrates product of matrices.

#include<iostream>
using namespace std;
int main()
{
int S[2][3] = {1,2,3,4,5,6 };
int P[3][1] = { 6,5,4 };
int C[2] = {0} ;

for ( int i =0 ; i<2; i++)

for ( int j = 0; j<1; j++)
for (int k = 0; k<3; k++)
❖ 194 ❖ Programming with C++
{ C[i] += S[i][k] * P[k][j] ;}

for ( int m =0 ; m<2; m++)

cout<<C[m] <<“\n”;
return 0;
}

The expected output is as below.

28
73

8.12 THREE DIMENSIONAL ARRAYS (ARRAYS OF MATRICES)

A 3-dimensional array can be taken as an array whose elements are two dimensional arrays. In
practice it may be taken as an array of matrices. An array with int elements may be declared as
below.
int A[m][n][p];
The following program illustrates the input and output of a three dimensional array.
PROGRAM 8.25 – Illustrates input / output of three dimensional arrays.

#include <iostream>
using namespace std;
int main()
{
int bill[2][3][4];
cout << “Write 24 integers separated by white space ;\n”;

//you may write in any groups, the program will read left to right.

for(int i=0;i<2;i++)
{for(int j=0;j<3;j++)
{for (int k = 0; k<4;k++)
cin>>bill[i][j][k];
}
}
cout<<“\n\n”;

for(int n=0;n<2;n++)
{for(int j=0;j<3;j++)
{for (int k =0;k<4;k++)
cout<<“bill[”<<n<<“][”<<j<<“][”<<k<<“] = ”<< bill[n][j][k];
cout<<“\n”; }}
cout<<“\n\n”;
 Arrays ❖ 195❖
for(int m=0;m<2;m++)
{for(int p=0;p<3;p++)
{for (int k =0;k<4;k++)
cout<< bill[m][p][k];
cout<<“\n”;}
cout<<“\n”; }
return 0;
}

The output is given below.

Write 24 integers separated by white space ;
111122223333444455556666

bill[0][0][0] = 1 bill[0][0][1] = 1 bill[0][0][2] = 1 bill[0][0][3] = 1

bill[0][1][0] = 2 bill[0][1][1] = 2 bill[0][1][2] = 2 bill[0][1][3] = 2

bill[0][2][0] = 3 bill[0][2][1] = 3 bill[0][2][2] = 3 bill[0][2][3] = 3

bill[1][0][0] = 4 bill[1][0][1] = 4 bill[1][0][2] = 4 bill[1][0][3] = 4
bill[1][1][0] = 5 bill[1][1][1] = 5 bill[1][1][2] = 5 bill[1][1][3] = 5
bill[1][2][0] = 6 bill[1][2][1] = 6 bill[1][2][2] = 6 bill[1][2][3] = 6

1111
2222
3333

4444
5555
6666

The output has also been tabulated in the form of two 3×4 matrices.

EXERCISES
1. How do you declare a single dimensional array?
2. How can array be initialized?
3. What will happen if the initialization has fewer elements than the number of elements of an
array?
4. What will be the output if the initialization has more number of elements than the declared
number of elements of an array?
5. How do you initialize a two dimensional array?
6. How would you access the first and last elements of the array A[6]?
❖ 196 ❖ Programming with C++
7. How do you initialize a three dimensional array?
8. Write a user interactive program to record 5 weights in floating point numbers as an array with
name Weights?
9. How do you determine the address of an array?
10. How do you determine the address of each element of an array?
11. Write code for sorting a list of integers in descending order.
12. Write code for sorting the following array in ascending order, i.e. A to Z.
char Myarray [10] = { ‘B’, ‘A’, ‘H’, ‘T’, ‘R’, ‘D’, ‘Z’, ‘S’, ‘N’, ‘L’};
13. Write code for determining the maximum value in an array of integer numbers and to determine
the subscript value of array element with the maximum value.
14. Write code for determining the array element with the minimum value in an array of floating
point numbers.
15. Write code for determining maximum value and minimum value in an array of integers. It should
also determine the corresponding array subscripts.
16. What is the physical interpretation of a 3-dimensional array?
17. What is typedef and how it is coded?
18. Make a program which reads a sample of 11 values in floating point numbers entered by user
and calculates the average value and standard deviation of the sample.
19. Make a user interactive program in which user is asked to enter four integers as elements of
an array, and displays them on the monitor.
20. Make a program in which a function is defined to return the square of elements of an array of
integers.
21. Write a program in which a function is defined to return the dot product of two vectors A and
B having 3 components each, i.e. A1, A2, A3 and B1, B2 and B3. The magnitudes of vector
components are integers which are entered by the user of the program.
22. Make a program to determine the average value and standard deviation of a sample of 10
measurements of diameters of jobs in floating point numbers and entered by user.
23. Make a program which reads 4 names typed and entered by a user and displays the names
on the monitor.
24. Make program which asks the user to enter 5 quotations for price of an item in floating point
numbers. The program calculates the average of the five and displays it on monitor.
25. For the following two dimensional array, write a program to carry out the output in rows and
columns.
Myarray[3][4] = { 1,2,3,4,11,12,13,14,21,22,23,24};
26. Make a program to carry out the product of two matrices.
27. Make a program to carry out the addition and subtraction of two matrices.

❍❍❍
CHAPTER

9.1 INTRODUCTION

Pointer is a variable which holds (or whose value is) the memory address of another variable. As
already discussed in Chapter1, the basic unit of memory is a bit. The bits are grouped into bundles
of eight each. These bundles are called bytes. The main memory (RAM) of computer consists
of bytes which are numbered sequentially. Each byte is numbered and this number is its address.
When we declare a variable we mention its type and its name, and the compiler along with
operating system allocates a block of memory and location for storing the value of the variable.
The number of bytes allocated to a variable for storing its value depends on its type. For example,
generally 4 bytes are allocated for storing the value of an int (integer). The address of a variable
is the address (byte number) of the first byte of the memory block allocated to the variable.
Thus the variable gets an address where its value is stored. You may say it is the house number
where the variable resides in computer memory. We can dig out the memory address of a variable
by using address-of operator (&) also called reference operator. For instance, if n is a variable,
its address is given by &n. We may declare another variable for storing &n which is the address
of n. That variable is called pointer to n.

9.2 DECLARATION OF POINTERS

Declaration of a pointer variable is done by mentioning the type of the variable for which it is
the pointer, followed by indirection operator, which is also called dereference operator (*),
and the name of the pointer variable. The operator (*) tells the compiler that the variable being
declared is a pointer. Name of a pointer variable may be decided as it is done for any other
variable. The illustrations of pointer declarations for int variables are given below.
int m, n,; // m and n are integer variables
int *ptr ; // ptr is pointer to any integer number
int* ptr = &n ; // ptr now points to n.
int *ptr = &m; // ptr now points to m
int* ptr = 0; // now ptr is initialized to 0.
Also note that it is immaterial whether pointer is declared as int* ptr or as int *ptr, i.e. the
asterisk (*) is attached to int in first case and to ptr in the second case. It is, however, a good
programming practice to initialized the pointers when they are declared. Pointers may be initialized
❖ 198 ❖ Programming with C++

to an address or otherwise to 0 or NULL which is a symbolic constant defined in C and inherited

by C++. NULL is also defined in <iostream> and other C++ header files. It is equivalent to 0.
A NULL pointer does not point to any valid data.
Here in this chapter, we shall deal with pointers to variable, arrays, strings and functions.
Pointers to classes, class members and class objects, etc., are discussed in Chapter 12.
A pointer to a variable with value in floating point decimal number such as PI = 3.14159;
and to a character variable may be declared as illustrated below. Let pf be the name of pointer
to a float variable, pd be the name of pointer to a double variable and pc be the name of pointer
to a character variable. These are defined below and illustrated in Fig.(9.1).
float
float PI = 3.14159; points to PI
pf 3.14159
float *pf = &PI;
char
char ch = ‘A’; points to ch
pc A
char *pc = & ch;
double
double D = 4328.6543 points to D
pd 4328.6543
double * pd = &D ;

Fig. 9.1: Pointers to different data types

The pointers to two or more variables of same type may be declared in the same line as carried
out in the following code.
int n,y , *ptrn, *py; // n and y are integers variables
// ptrn and py are pointers
ptrn = &n; // initialization of pointer ptrn by &n
py =&y; // initialization of pointer py by &y
In the above declarations, ptrn is pointer to n and py is pointer to y. We may also use typedef
as illustrated below to declare a number of pointers of same type.
typedef double* ptd;
double x,y,z;
ptd px = &x, py = &y, pz = &z ;

INDIRECTION OR DEREFERENCE OPERATOR

The value of a variable may be obtained from its pointer by using indirection operator also called
dereference operator (*). It is called indirection because it obtains the value indirectly. An
application of dereference operator is given below.
int n = 60, *ptrn ;
ptrn = &n ;
*ptrn = 60; // use of dereference operator.
Program 9.1, given below, illustrates the declaration of pointers and application of address-
of operator (&). A pointer is also a variable and is allocated a block of memory for storing its value
 Pointers ❖ 199❖

which is address of another variable. The address-of operator (&) may also be used to get the
address of the pointer.

PROGRAM 9.1 – Illustrates declaration of pointers and use of address-of operator (&)

#include <iostream>
using namespace std;
int main()
{ int n = 60 ;
double B = 8.56;

int* ptrn = &n; //ptrn is pointer to int variable n.

double *pB = &B; //pB is pointer to a double variable B
cout<<“n = ”<<n<<“, B = ”<< B<<endl;
cout<<“The address of n is = ”<<&n<<“ , ptrn = ” << ptrn <<endl;
cout<< “*ptrn = ”<<*ptrn<< “, *pB = ” <<*pB <<endl;
cout <<“The address of B = ” << &B << “, pB = ” << pB << endl;
cout << “Address of ptrn = ” <<& ptrn<<endl;
return 0;
}

The expected output is given below.

n = 60, B = 8.56
The address of n is = 0012FF7C , ptrn = 0012FF7C
*ptrn = 60, *pB = 8.56
The address of B = 0012FF74, pB = 0012FF74
Address of ptrn = 0012FF70

From the output of the above program it is clear that n and *ptrn have the same value, i.e.
60. Similarly B and *pB are both equal to 8.56. Thus it shows that use of dereference operator
(*) on pointer variables ptrn and pB gives the values stored at the respective addresses. The program
also gives the address where ptrn is stored. Here we have attached ptr or p to the name of variable
to indicate that it is a pointer. But you can put any other name which is legal in C++. The
following program illustrates some more pointer declarations and the use of typedef.
PROGRAM 9.2 – Illustrates use of typedef in pointer declaration.

#include <iostream>
using namespace std;
int main()
{char ch = ‘S’;
char* pc = &ch;
cout << “*pc = ”<< *pc<<endl;
double x = 4.5, z;
typedef double* pd;
❖ 200 ❖ Programming with C++
pd px = &x; // here pd stands for double*
cout<<“px = ”<<px<<endl; //display value of px
cout <<“&x = ” << &x<<endl; // display address of & x
pd pz = &z;
pz = px; // px (address of x) is assigned to pz
cout << “*pz = ” <<*pz<<endl;
cout<< “*px = ”<<*px <<endl;
return 0;
}

The expected output is given below. Output shows that address of x is same as value of px
and by equating pz = px we make pz also to point to value of x as illustrated below in Fig. 9.2.

*pc = S pz
px = 0012FF70 double
&x = 0012FF70 4.5 x
*pz = 4.5
*px = 4.5 px

Fig. 9.2: Two pointers pointing to same value

The pointers to all types of data occupy the same size of memory because value of a
pointer is the memory address – an unsigned integer. However, the variables to which the
pointers point to, of course, occupy different sizes of memory blocks according to their types.
This is illustrated in Program 9.3 given below. From the output of the program we find that
the sizes of memory blocks allocated to pointers for different types of variables are equal to 4
bytes.
On most of 32 bit systems, a pointer is allocated 4 bytes of memory space, (see the following
program)

PROGRAM 9.3 – Illustrates that size of pointers to all data types is same.

#include <iostream>
using namespace std;
int main()
{ int n = 60,*pn;
double PI = 3.14159 ,*pPI;
pPI = &PI;
pn = &n;
char ch = ‘M’,*sam; // sam is the name of pointer
sam = &ch; // initializing by &ch.
cout<< “Size of the pointer to int n = ” <<sizeof(pn)<<endl;
cout <<“Size of the pointer to double PI = ”<<sizeof(pPI)<<endl;
 Pointers ❖ 201❖
cout<< “ Size of the pointer to char ch = ”<<sizeof(sam) <<endl;
cout<<“Variables are ”<<*pn <<“, ”<<*pPI<<“ and ”<<*sam<< endl;
return 0;
}

The expected output of the program is given below.

Size of the pointer to int n = 4
Size of the pointer to double PI = 4
Size of the pointer to char ch = 4
Variables are 60, 3.14159 and M
The output shows that size of memory allocated to a pointer is 4 bytes irrespective of the
fact whether the variable to which it points to is integer, double or character. Program 9.4 given
below also illustrates that actions of dereference operator and that of address-of operator (&) are
opposite to each other.

PROGRAM 9.4 – Illustrates the address-of operator (&) and dereference operator (*)

#include <iostream>
using namespace std;
main()
{ int n = 60;
int* ptrn , *ptrm ; //declaration of pointers to int
ptrn = &n;
ptrm = ptrn; // assigning a pointer
cout<<“n = ”<<n<< “\t&n = ”<<&n <<“\t *&n = ”<< *&n<<endl;
cout<< “ptrn = ” <<ptrn<< “ “ << “ptrm = ” << ptrm <<endl;
cout<< “The value pointed to by ptrn = ” << *ptrn<<“\n” ;
// *ptrn is the value pointed to by pointer ptrn
cout << “The value pointed to by ptrm = ” <<*ptrm <<endl;
cout << “Address of ptrn = ” <<&ptrn<<endl;
double PI = 3.141592;
double *pPI; // declaration of pointer for PI
pPI = &PI;
cout<<“PI=”<<PI<<“,\t &PI = ” <<&PI<<“\n”;
cout <<“*&PI = ” <<*&PI<< “ ,\t *pPI = ” << *pPI<<“\n”;
return 0;
}

The expected output is given below.

n = 60 &n = 0012FF7C *&n = 60
ptrn = 0012FF7C ptrm = 0012FF7C
The value pointed to by ptrn = 60
The value pointed to by ptrm = 60
❖ 202 ❖ Programming with C++

Address of ptrn = 0012FF78

PI=3.14159, &PI = 0012FF6C
*&PI = 3.14159 , *pPI = 3.14159

From the output of the above program it is clear that n is an integer with value 60. The
address of n is 0012FF7C. The value of ptrn which is pointer to n is also 0012FF7C. The value
of ptrn is assigned to ptrm to creat another pointer to x. The address of ptrn can also be
determined by the operator &. Thus the address of ptrn is determined as 0012FF78 which is 4
bytes away from 0012FF7C. The value of variable n may be determined by dereference operator
(*). It is determined in the first line of output as 60. The memory blocks allocated for storing
the value of n and its pointer are illustrated in Fig. 9.3a.
The second part of the output of the program similarly deals with a double number PI=
3.14159. The pointer for PI has been given the name pPI. The address of PI is 0012FF6C. The
value of PI may be obtained by calling PI or by *&PI or by *pPI. This shows that actions of
operators * and & are opposite to each other. The action of dereference operator (*) is illustrated
in Fig. 9.3b.
n ptrn

RAM memory
60 0012FF7C

Byte Number. 0012FF7C 0012FF78 0012FF74

Fig. 9.3a: An integer and its pointer in memory

Ptrm = ptrn; Ptrm *Ptrm n

0012FF7C

60
Ptrn *ptrn
0012FF7C

Fig. 9.3b: Two pointers pointing to same value

9.3 PROCESSING DATA BY USING POINTERS

The pointers may be used in place of the variables they point to. We have already seen in the
above program that n and *ptrn have the same value. Now if there is a change in value of *ptrn
the n will also change accordingly.
The following program computes the circumference and area of a circle having diameter D.
Instead of D and PI = 3.14159, we make use of pointers to D and PI. The value of PI is given
in the program while the diameter value is entered by the user. The two functions, i.e. area and
circumference are written in terms of pointers.
 Pointers ❖ 203❖

PROGRAM 9.5 – The program illustrate the application of pointers in computations.

#include <iostream>
using namespace std;
int main()
{ int D , *ptrD = &D ; //declaration of D and pointer to D
double PI= 3.14159;
double* ptrPI = &PI; //pointer to PI
cout<<“Write the diameter of the circle: ”;
cin>>*ptrD;

cout<<“Diameter = ”<<*ptrD <<endl;

double Circumference = *ptrD * *ptrPI; // = PI*D
double Area = *ptrPI* *ptrD **ptrD/4; // = PI*D *D /4
cout<<“Circumference = ” <<Circumference <<endl;
cout<< “Area of circle = ” << Area<<endl;
return 0;
}

The expected output of the program is as under.

Write the diameter of the circle: 20
Diameter = 20
Circumference = 62.8318
Area of circle = 314.159

9.4 POINTER TO POINTER

Just like we have a pointer to a variable we can also define a pointer to a pointer which keeps
the address of the pointer to the variable. In the following program, n is an integer variable with
value 57, ptrn is the name of pointer to n and pptrn is the name of pointer to ptrn. The pointer
ptrn to a variable n and pointer to pointer pptrn are declared and initialized as illustrated below.
int* ptrn = &n ; // ptrn is pointer to n
int** pptrn = & ptrn ; // pptrn is pointer to ptrn
Note that both for pointer and pointer to pointer the type is int because n is integer. Also
note that for getting the value of variable we use one dereference operator (*) to pointer while
for getting the value of variable from pointer to pointer, we have to use two dereference operators
(**). But there is no change in the application of address-of operator (&). Also note that we
cannot use &&n because & requires an l-value whereas &n is a r-value. The object that is placed
on left side of assignment operator ( = ) is l-value and the one placed on right side is r-value.
❖ 204 ❖ Programming with C++

PROGRAM 9.6 – The program illustrates operations with pointer to pointer.

#include <iostream>
using namespace std;
int main()
{ int n = 57 ;
int* ptrn = &n; //Pointer to n
cout<<“n = ”<<n <<“, Address of n = “ <<&n <<endl;
cout<<“ptrn = ”<< ptrn<<endl;
cout << “Address of ptrn = ” << &ptrn<<endl<<end1;
int** pptrn = &ptrn; // Pointer to the pointer to n
cout << “pptrn = ”<<pptrn<<endl;

int*** ppptrn = &pptrn; //Pointer to pointer to pointer to n

cout << “Address of pptrn = ” << &pptrn <<endl;
cout << “ppptrn = ” << ppptrn <<“\n”;
cout<<“*ptrn = ”<<*ptrn<<endl;
cout<<“*pptrn = ” << *pptrn<<endl;
cout<<“**pptrn = ” << **pptrn<<endl;
cout << “***ppptrn = ” << ***ppptrn <<endl;
return 0;
}

The output is given below.

n = 57, Address of n = 0012FF7C
ptrn = 0012FF7C
Address of ptrn = 0012FF78

pptrn = 0012FF78
Address of pptrn = 0012FF74
ppptrn = 0012FF74
*ptrn = 57
*pptrn = 0012FF7C
**pptrn = 57
***ppptrn = 57
The above output is self explanatory and is partly illustrated in Fig. 9.4 given below.
 Pointers ❖ 205❖

Pointer to pointer to pointer Pointer to pointer Variable

Pointer
Name ppptrn pptrn ptrn n
*ppptrn *pptrn *ptrn
Value
0012FF74 0012FF78 0012FF7C 57

Memory address
0012FF74 0012FF78 0012FF7C
**pptrn
***ppptrn

Fig. 9.4: The numbers written within the memory blocks are the values stored in the memory
blocks. The addresses of the blocks are written below the blocks
In the declaration of pointers to floats and double objects the two types should not be
intermixed as it is illustrated below.
float m ; // variable of type float
float* ptrm = & m; // pointer to float, do not use double.
float** kptrm = & ptrm; // pointer to pointer, type float
double n ; // variable double
double* ptrn = &n; // pointer to double, do not use float.
double** kptrn = &ptrn // pointer to pointer to double
Do not mix float and double in pointers. Rather do not mix types when dealing
in pointers.

CONSTANT POINTERS
The attribute modifier const may be used with pointers as well. However,
the declaration should be done carefully keeping in view the following.
const int* ptr ; // Here ptr is a non-constant pointer to a constant int
int* const ptr ; // Here ptr is a constant pointer to a non-constant int.
const int* const ptr ; // Here ptr is a constant pointer to a constant int.

9.5 POINTERS AND ARRAYS

When we declare an array, its identifier (name) is a const pointer to the first element of array. The
pointer to array also carries the address of first element of array. Thus the pointer to an array has
same value as the name of the array. The difference is that the name of an array is a const pointer,
i.e. it cannot be incremented or changed while the pointer to an array which carries the address of
first element of the array may be incremented. If pointer to an array is incremented by 1 it would
point to the next element of the array. Thus the value of the pointer would increase by the number
of bytes allocated to one array element. The declaration of a pointer to an array is illustrated below.
int A[] = {12, 25, 36 ,50}; // A is the name of array
int *ptrA ;
ptrA = A ; // Notice there is no & operator before A
The above definition may also be written as below.
int *ptrA = A;
and also as,
int* ptrA = &A[0];
❖ 206 ❖ Programming with C++

It is so because the address of A[0] is same as value of A. The elements of array are A[0], A[1],
A[2], and A[3] have respective values as 12, 25, 36 and 50 which may also be obtained from
pointers as illustrated below.
A[0] = *ptrA;
A[1] = *(ptrA+1);
A[2] = *(ptrA+2) ;
A[3] = *(ptrA+3);
The above discussion shows that array subscript and pointer offset are equal. If we call array
A without a subscript it will give the address of first element of array. A +1 gives the address of
second element, A+2 gives address of third elements and so on. Therefore, we can also get the
values stored at these addresses by using dereference operator (*). Thus for the above declared
array we may get the values of array elements as given below.
*A = 12
*(A+1) = 25
*(A+2 ) = 36
*(A+3) = 50
Note that expression *A++ is not legal because A is a constant pointer to array and a constant
cannot be incremented. Also note that array subscripts are equal to pointer offset in traversing
of array elements.

PROGRAM 9.7 – Illustrates declaration of pointer to an array.

#include <iostream>
using namespace std;
int main()
{ int A [] = {80,40,20,10}; // A is an array of integers
int *ptrA = A; //ptrA is pointer to A. Note that & is not used.
cout <<“*ptrA = ”<<*ptrA <<“ , \tptrA = ”<<ptrA<<endl;
cout <<“*(ptrA+1) = ” << *(ptrA+1)<< “ ,\tptrA+1 = ” <<ptrA+1<<endl;
cout<<“*(ptrA+2) = ”<<*(ptrA+2)<< “ , \tptrA + 2 = ” <<ptrA+2 <<endl;
cout<<“*(ptrA+3) = ”<<*(ptrA+3)<< “ , \tptrA+3= ” << ptrA+3<<“\n\n”;
cout << “A = ” << A<<endl;
cout<< ptrA[0]<<“\t”<<ptrA[1]<<“\t”<<ptrA[2]<<“\t”<<ptrA[3]
<<endl;
cout << *(A+0) <<“\t”<<*(A+1) << “\t”<<*(A+2) << “\t” << *(A+3)<<endl;
// These will display the values of array elements.
return 0;
}

The expected output is as below.

*ptrA = 80 , ptrA = 0012FF70
*(ptrA+1) = 40 , ptrA+1 = 0012FF74
 Pointers ❖ 207❖
*(ptrA+2) = 20 , ptrA + 2 = 0012FF78
*(ptrA+3) = 10 , ptrA+3= 0012FF7C
A = 0012FF70
80 40 20 10
80 40 20 10

From the output given above note that value of A is same as that of pointer ptrA to the array.
Both give the address of first element of array. A = ptrA = 0012FF70.
Now ptrA+1 has the value 0012FF74 which is 4 bytes away from the address of first element
and at that address the value stored is *(ptrA+1) = 40; also *(A+1) = 40; which is the second
element of the array. Therefore, when a pointer to an array is incremented by 1 it points to the
next element in the array. Its value is, in fact, incremented by the number of bytes allocated to
one element of array. Similarly ptrA+2 is pointer to the third element, ptrA+3 is the pointer to
the fourth element of the array. The following program illustrates traversing of array with
pointers.

PROGRAM 9.8 – Illustrates traversing an array by pointers.

#include <iostream>
using namespace std;
int main()
{ int kim [ ] = {50, 86, 90, 12} ;

int* pkim = kim; // defining the pointer to array kim

for ( int i=0 ; i<4;i++) // for loop for output of array
cout<<“kim[”<<i<<“] = ” << *(pkim +i) <<“ , ”;
//*(pkim +i)is the value of kim[i]element of array
return 0;
}

The expected output of the program is given below.

kim[0] = 50 , kim[1] = 86 , kim[2] = 90 , kim[3] = 12 ,

In case of arrays the increment in pointer by unity, in fact, increases the value of pointer equal
to size of one array element, thus if pkim points to first element of array, pkim+1 points to the
second element of array. In the following program the array elements are subjected to arithmetic
operations with the help of pointers.

PROGRAM 9.9 – Manipulating array elements with pointers.

#include <iostream>
using namespace std;
int main()
{
❖ 208 ❖ Programming with C++
int kim [ ] = {6,8,5,4} ; //4 element array
int* pkim = kim; // Declaration of pointer
for ( int i=0 ; i<4;i++)
cout<<“kim2[”<<i<<“] = ”<< *(pkim +i)**(pkim +i) <<“, ”;
// creating kim2 array from elements of kim
cout <<“\n”;
for(int j= 0; j<4; j++) // creating kim3 array
cout<<“kim3[”<<j<<“]= ”<<*(pkim+j)**(pkim +j)–*(pkim+j)<<“, ”;
return 0;
}

The expected output is as below. The output is explained by comments.

kim2[0] = 36, kim2[1] = 64, kim2[2] = 25, kim2[3] = 16,
kim3[0]= 30, kim3[1]= 56, kim3[2]= 20, kim3[3]= 12,

9.6 ARRAY OF POINTERS TO ARRAYS

In the following program an array of pointers int *pAB[2] is declared for two arrays. The first
element of the pointer array, i.e. pAB[0] points to array A with four elements 80, 70, 60, 50
and the second element, i.e. pAB[1] points to array B with elements 10, 20, 30 and 40. The
program illustrates how to carry out the output of the elements of A and B.

PROGRAM 9.10 – Illustrates an array of pointers to arrays.

#include <iostream>
using namespace std;
int main()
{ int A [] = {80,70, 60,50} ; // Array A
int B [] = {10, 20, 30, 40}; // Array B
int* pAB[] = {A, B} ;
//Array of pointers initialized by names of arrays A and B
cout<< “Address of A = ” << A <<endl;
cout << “Address of B = ” << B <<“\n\n”;
cout << “Address of first element of pAB[0] = ”<< pAB <<endl;
cout << “Address of second element of pAB[1] = ”<< pAB +1<<endl;
cout <<“*pAB = ”<<*pAB <<“,\t**pAB = ”<<** pAB <<“\n\n”;
// Here *pAB = A and **pAB = A[0], see the output below.
cout <<“A[0] = *(*pAB) = ”<< * (* pAB) << endl;
cout<<“A[1] = *(*pAB+1) =” << *(*pAB+1)<< endl;
cout<<“A[2] = *(*pAB+2) = ”<< *(*pAB +2) << endl;
cout<<“A[3] = *(*pAB+3) = ”<<*(*pAB +3)<<endl<<endl;
 Pointers ❖ 209❖

cout<<“*(pAB+1)= ” <<*(pAB+1) <<“\t **(pAB+1) = ” <<**(pAB+1)<< endl;

// Here *(pAB+1) = B and **(pAB +1) = B[0], see output.
cout<< “B[0]= *(*(pAB+1)) = ”<<*(*(pAB+1)+0) << endl;
cout<<“B[1] = *(*(pAB+1)+1) = ”<<*(*(pAB+1)+1)<<endl;

cout<<“B[2]= *(*(pAB+1)+2) = ”<<*(*(pAB+1)+2)<<endl;

cout<<“B[3] = *(*(pAB+1)+2) =”<<*(*(pAB+1)+3 )<<endl;
return 0;
}

The expected output is given below. If you carefully go through the output it is self
explanatory.
Address of A = 0012FF70
Address of B = 0012FF60

Address of first element of pAB[0] = 0012FF58

Address of second element of pAB [1] = 0012FF5C
*pAB = 0012FF70, **pAB = 80

A[0] = *(*pAB) = 80
A[1] = *(*pAB+1) =70
A[2] = *(*pAB+2) = 60
A[3] = *(*pAB+3) = 50

*(pAB+1) = 0012FF60 **(pAB+1) = 10

B[0]= *(*(pAB+1)) = 10
B[1] = *(*(pAB+1)+1) = 20
B[2]= *(*(pAB+1)+2) = 30
B[3] = *(*(pAB+1)+2) =40

In the code of the above program you would notice that initialization of pointer array has
been carried out by array names because they are constant pointers to the arrays. In the following
program we have three arrays and we declare an array of pointers for them. The initialization of
pointer array is carried out by the names of arrays.
PROGRAM 9.11 – Another illustration of array of pointers to arrays.
#include <iostream>
using namespace std;
int main()
{ int Kim [] = {6,8,5,4} ; // 4 element array
int Bill [] = {2, 3, 7 }; // 3 element array
int Sim[] = {10, 11, 12, 13,14 }; //5 element array

int* pKBS [] = { Kim , Bill, Sim};

// Array of pointers, initialization done by array names
cout<<“Kim : ”<<**pKBS<<“\t”<<*(*(pKBS)+1) <<“\t”<<*(*(pKBS)+2) << “\t”
<< *(*(pKBS)+3)<<endl; // Getting elements of Kim
❖ 210 ❖ Programming with C++
cout<<“Bill : ”<<*(*(pKBS+1))<<“\t”<<*(*(pKBS+1)+1)<<“\t”<< *(*(pKBS+1)+2)
<<endl; // Getting elements of Bill
cout<<“Sim : “ << *(*(pKBS+2))<< “\t” << *(*(pKBS+2)+1) << “\t” <<*(*(pKBS
+2)+2) <<“\t” <<*(*( pKBS+2)+3)<< “\t” << *(*( pKBS+2)+4) <<endl;
// Getting elements of Sim
return 0 ;
}

The expected output is given below.

Kim : 6 8 5 4
Bill : 2 3 7
Sim : 10 11 12 13 14

9.7 POINTERS TO MULTI-DIMENSIONAL ARRAYS

POINTERS TO TWO DIMENSIONAL ARRAY

Program 9.11 is in fact an example of two dimensional array, because, an array whose elements
are one dimensional array is a two dimensional array. In declaration of pointer to two dimensional
array also we assign the address of array by its name which is the address of the first element of
the array. The declaration is done as follow.
int A[n][m] ; // A is a two dimensional array
int (*pA)[m] = A ; // pointer to two dimensional array
An element of two dimensional array with subscripts i and j may be accessed as given below.
A [i][j]
In pointer notation the element may be accessed by the following expressions.
*(*(A+ i) + j)
Or by a pointer pA as under.
*(*(pA +i)+j)

The following program illustrates the declaration and output of a two dimensional array.
PROGRAM 9.12 – Illustrates pointers to two dimensional arrays and output of arrays.

#include <iostream>
using namespace std;
int main()
{ int KBS [3][4] = {6,8,5,4, 2,3,7,9, 10,11,12,13};
int(*pKBS )[4] = KBS ; // declaration of pointer to KBS
cout<<**pKBS<<“\t”<<*(*(pKBS)+1)<<“\t”<<*(*(pKBS)+2)<<“\t”
<<*(*(pKBS)+3)<<endl;
cout<<*(*(pKBS+1))<<“\t”<<*(*(pKBS+1)+1)<<“\t”<<*(*(pKBS+1)+2)<<“\t”<<
*(*(pKBS+1)+3) <<endl;
cout<<*(*(pKBS+2))<<“\t”<<*(*(pKBS+2)+1)<<“\t”<<*(*(pKBS+2)+2)
<<“\t”<<*(*( pKBS+2)+3)<<endl;
return 0 ; }
 Pointers ❖ 211❖
The expected output is given below.
6 8 5 4
2 3 7 9
10 11 12 13

From the output it is clear that A[i][j] and in the pointer notation *(*(pA+i)+j) are equivalent
expressions. So we may as well use the expression like *(*(A+i)+j)for traversing a two dimensional
array or a matrix.

DECLARATION OF POINTER TO THREE DIMENSIONAL ARRAY

A three dimensional array is an array of matrices. Let A be a 3 dimensional array. The pointer
declaration and initialization is carried out as below.
int A[2][3][4] ;
int (*pA)[3][4] = A ; // pointer to array A
Here (*pA ) [3][4] is the pointer to A[2][3][4]. An element of array with subscripts values
i, j and k may be accessed by the following expression.
A[i][j][k]
And in pointer notation an element with subscripts i, j and k may be accessed by expression
given below.
*(*(*(A +i) + j) + k)
In the following program a 3 dimensional array is declared. Its output is carried out in pointer
notation as an array of two matrices.

PROGRAM 9.13 – Illustrates pointer to three dimensional array and array output.

#include <iostream>
using namespace std;
int main()
{int KBS [2][3][4] = {6,8,5,4, 2,3,7,9, 1 ,2,3,4, 21,22,23,24, 31,32,33,34,
41, 42, 43, 44} ; // 3 dimensional array
int(* pKBS )[3] [4] = KBS; // Declaration of a pointer
for (int i =0; i<2; i++)
{for ( int j =0; j <3; j++)
{for ( int k =0; k<4; k++)
cout << *(*(*(pKBS +i) + j) +k)<<“ ”;
cout<<“\n”;}
cout<<“\n”;}
return 0 ;
}
❖ 212 ❖ Programming with C++

The expected output is given below.

6 8 5 4
2 3 7 9
1 2 3 4

21 22 23 24
31 32 33 34
41 42 43 44

9.8 POINTERS TO FUNCTIONS

In the previous section we have seen that name of an array carries the value of address of first
element of array and is a constant pointer to the array. Similarly the name of a function also carries
the address of function and is a constant pointer to the function. Therefore, we can declare a
pointer to a function and initialize it with the name of the function. The pointer to a function
may be declared as below.
type ( *identifier_for_pointer ) ( types_of_parameters_of_ function);
Here type refers to the type of data returned by the function, in the first parentheses is the
indirection operator followed by name of pointer. The next pair of parentheses contains the types
of parameters. Let Fptr be name of pointer to a function Func of type int and which takes two
integer parameters, i.e. int Func(int, int). The pointer to this function is declared as,
int (*Fptr) (int, int) ;
The above declaration is, in fact, for any function which has return value of type int and has
two int parameters. Let int Func1(int, int) and int Func2(int, int) be any two functions which
satisfy the above criteria. The above declaration of pointer applies to both these functions. Hence,
if we may assign the following value to Fptr,
Fptr = Func1; // Fptr points to Func1
The pointer Fptr points to Func1. The function Func1 may be called by dereference operator.
int a , b;
(*Fptr)(a, b ); // calling the function
In fact the dereference operator is not required and function may simply be called as below.
Fptr(a, b);
Now we may as well initialize Fptr with address of Func2.
Fptr = Func2; // Pointer Fptr now points to Func2
With the above definition, the pointer Fptr now points to Funct2. The function Funct2 may
be called by the expression (*Fptr)(i,d); or by Fptr(i,d); where i and d are integers. This gives the
facility to the user of the program to choose any one of the functions provided in the program
that matches the pointer declaration and user’s requirement.
In the above declarations *Fptr has been enclosed in parentheses because otherwise there
would be only one parentheses that encloses the parameter, and which will have higher precedence
to (*) operator. Thus an expression such as int *Fptr(int, int); would mean that function Fptr
takes two int arguments and returns a pointer to an integer. Therefore, it is necessary to enclose
*Fptr in parentheses. The following program illustrates the use of pointers to functions. The user
has a choice of four functions and choice is facilitated by switch statement.
 Pointers ❖ 213❖
PROGRAM 9.14 – Illustrates pointers to functions.

#include <iostream>
using namespace std;
int F1(int n){ return n ;} // returns argument
int F2( int n){ return n*n;} // returns square of argument
int F3(int n ){ return n*n*n;} // returns cube of argument
int F4( int n){ return n*n*n*n;} //returns 4th power of argument

int main()
{ int m;
int (*Fptr)(int m); // pointer to above functions
cout << “Writ an integer number “;
cin >> m ;
int userchoice ; // For user to choose the function
cout << “Give your choice between 1 to 4 :”;
cin >> userchoice;
switch (userchoice) // switch condition for multiple choice
{
case 1:
Fptr = F1 ; // selects function F1
cout << m << “raised to power 1 = ” << F1(m)<< endl ;
break;
case 2 :
Fptr = F2 ; // selects function F2
cout << m << “raised to power 2 = ” << F2(m)<<endl;
break;
case 3 :
Fptr = F3 ; // selects function F3
cout << m << “raised to power 3 = ” << F3(m)<<endl;
break;
case 4 :
Fptr = F4 ; // selects function F4
cout << m << “raised to power 4 = ” << F4(m)<<endl;
break;
default :
cout << “Make a second choice between 1 to 4 “<<endl;
}
return 0;
}

The expected output to first choice with m = 5 and user choice equal to 3.
❖ 214 ❖ Programming with C++
Writ an integer number 5
Give your choice between 1 to 4: 3
5 raised to power 3 = 125
The second trial was done with m = 6 and user choice = 5 which is not there, so it goes to
default.
Writ an integer number 6
Give your choice between 1 to 4: 5
Make a second choice between 1 to 4

9.9 ARRAY OF POINTERS TO FUNCTIONS

Functions cannot be elements of an array, however, we can have an array of pointers which point
to different functions. The functions may be called in the main ( ) through the pointers. In the
following program we declare an array of pointers to three functions which are of type int. The
functions are defined out side the main ( ) , because a function cannot be defined in side another
function. The first function returns the argument, the second returns square of argument and
third returns the cube of argument.

PROGRAM 9.15 – Illustrates an array of pointer to functions.

#include<iostream>
using namespace std;

int func1(int); // function prototype of Func1

int func2(int) ; // function prototype of Func2
int func3(int); // function prototype of Func3

int main()
{
int P;
int (*F[3])(int)={func1,func2,func3}; // Array of three pointers
// initialization done by names of functions

cout << “Enter a number ”;

cin>> P;
cout<<“You entered the number ”<<P<<endl;
for(int i=0;i<3;i++) //for loop for calling different functions.
cout<< “Power ”<<i+1<<“ of the number = ”<<(*F[i])(P)<<endl;
return 0;
}
int func1(int a) // Definition of func1
{return a ; }

int func2(int a) // Definition of func2

{return a*a; }

int func3(int a) // Definition of func3

{return a*a*a;}
 Pointers ❖ 215❖

The for loop used in the above program, selects different functions which return the
corresponding values. These are given in the output of the program which is given below.
Enter a number 5
You entered the number 5
Power 1 of the number = 5
Power 2 of the number = 25
Power 3 of the number = 125
One of the applications of pointers to functions is in interactive programming in which the
user has a choice to make. The different choices can lead to different subroutines to execute the
choice.

9.10 POINTER TO FUNCTIONS AS PARAMETER OF ANOTHER FUNCTION

The name of a function holds the address of the function. A function may have another function
as one of its parameters. Thus a function may be passed on to another function as its argument
either by its name or by its pointer which also holds the address of the function parameter. With
pointer to function as parameter there is facility of choice of functions. And we can get more
than one value out of a function.
In the following program a function by name Function is of type double, i.e. returns a double
number. Pointer to other functions, i.e. double(*)(double, double, double), is one of its
parameters. The other three parameters are double, double, double. The pointer is used to call
two functions, i.e. volume and surface area of a cubical prism.

PROGRAM 9.16 – Illustrates pointer to functions as a parameter of a function.

#include <iostream>
using namespace std;
double Function(double (*)(double, double, double ),double, double, double);
// prototype of Function ()
double volume(double, double, double); // Function prototype
double surface(double, double, double ); // Function prototype
int main()
{ cout<<“Surface area and volume of a cubiod of sides 4.0,5.0,6.0:”<<endl;
cout<<“Surface area = ” <<Function(surface , 4.0,5.0,6.0)<<endl;
cout<< “Volume = ” <<Function(volume,4.0,5.0,6.0) <<endl;
return 0;
}
// definition of functions
double Function(double(*pF)(double k, double m, double n ),double p, double
q, double r) // Definition of Function
{ return (*pF)(p,q,r) ; }
❖ 216 ❖ Programming with C++
// Below is definition of function surface
double surface(double k, double m, double n)
{ return 2*(m*k+ n*k+ m*n) ; }
// Below is definition of function volume
double volume(double k, double m, double n)
{return k*m*n; }

The output of the program is given below.

Surface area and volume of a cubiod of sides 4.0,5.0,6.0:
Surface area = 148
Volume = 120

9.11 THE new AND delete

In many implementations of programs it is not known how much memory would be needed
for the successful execution. In such cases memory has to be allocated during the execution of
program. Such an allocation of memory is called dynamic allocation. In C language the function
malloc() is used for allocation of memory and function free() (header file <stblib.h>) is used to
free the memory so allocated. In C++ the operators new and delete are defined in the header
file <new> for the same purpose. However, just for memory allocation we need not include the
header file <new> in a program. But for special operators such as placement new the header file
has to be included. The operator placement new enables you to create an additional object at
predetermined memory location. For arrays, the new[] and delete[]are defined. If new[] is used
for creation of a new array, then delete[]should be used for its removal. If only delete is used it
will remove only the first element of the array and would result in memory leak.
The operator new allocates memory for the additional object and returns pointer to it. For
variables of types int, float, double and char the application of new is illustrated below. However,
new can also be used for objects of user defined types as well. The applications for user defined
types are dealt in Chapter 12. Declaration of a pointer for a new variable may be done as below.
int* k ; // k is pointer to an integer
k = new int; //k is pointer to an integer
*k = 15; //assigning a value to new integer.
Note that in the above declaration name of variable is not declared, only pointer name is
declared. The value of variable is obtained by dereference operator. The declaration and
initialization may be done in the same code line as well. It is illustrated below.
int* k = new int(15);
double *d = new double (40.8);
char*pc = new char (‘H’) ;
For arrays we may use the operator new as illustrated below.
char* ch ;
ch = new char[6];
The following program illustrates the application of new and delete. Notice that variable
name is not used. The value of variable is obtained by dereference operator (*).
 Pointers ❖ 217❖
PROGRAM 9.17 – Illustration of application of new and delete for fundamental types.

#include <iostream>
using namespace std;
main()
{ int* k ;
k = new int; // k is name of pointer.
*k = 8; // assigning value to new integer but without a name
float *D ;
D = new float (4.0); // Here D is name of pointer
cout<< “Address of k = ” << k <<endl;
cout << “Area of circle of diameter 4.0 = ” << 3.14159**D**D/4 <<endl;
delete k;
delete D;
return 0;
}

The output is as given below.

Address of k = 004800D0
Area of circle of diameter 4.0 = 12.5664

The following program illustrate the application of new[] for creating new arrays.

PROGRAM 9.18 – Illustration of application of new [ ] and delete [ ].

#include <iostream>
using namespace std;

int main()
{
int* k ;
k = new int[5];
cout<< “Enter 5 integers : ” ;

for (int i =0;i<5;i++)

cin >> k[i]; // user input of array elements
char *ch ;

ch = new char[6] ;
cout<<“Enter a name of 6 characters: ”;

for (int m = 0 ; m<6;m++)

cin >> ch[m]; // ch is similar to name of array. So ch[m]
// represents element at index value m
cout<< “Address of k = ” << k <<endl;

for (int n = 0 ; n<5; n++)

cout << k[n]<<“ ”; // here k is similar to name of array
❖ 218 ❖ Programming with C++
cout<<“\n”;
for (int j = 0 ; j<6; j++)
cout << ch[j]<<“” ; // for output of array elements
cout<<“\n”;
delete []k; //notice the code for deletion of new array
delete []ch;
return 0;
}

The expected output of the program is as below.

Enter 5 integers : 4 5 6 7 8
Enter a name of 6 characters: MANALI
Address of k = 004900D0
4 5 6 7 8
MANALI

9.12 REFERENCES

Often people give alternative name to a person, may be out of love or it is simply easier to
pronounce. For example, the family members and friends of Parminder may call him Pummy.
Both Parminder and Pummy refer to the same person called Parminder. Here Pummy is a
reference to Parminder.
Reference for a variable is also an alias or just another name for the same variable. The variable
name and its reference-name both point to the same block of memory where the value of variable
is stored. You must note that any change in the value of reference also changes the value of
variable.
For creating a reference first write the type of the variable for which the reference is being
created then write reference operator (&) followed by the name or identifier for the reference.
Name of reference may be decided in the same way as it is done for any variable. For example,
let there be an integer by name n. Reference to n with name Count may be defined as illustrated
below.
int n = 100;
int &Count = n; // Count is declared a reference to n.
Here Count is a reference to n. Let Weight be a floating point variable. We may define a
reference to it with name W as illustrated below.
double Weight ;
double &W = Weight ;
Similarly the reference by name T for a char variable ch may be defined as
below.
char ch = ‘S’;
char & T = ch ;
 Pointers ❖ 219❖

In the following program we declare three variables. One integer by name n, second a double
by name Weight and third a char by name ch. They have the values 100, 35.6 and ‘S’ respectively.
For each of these variables we create a reference. The name of reference for n is Count, the name
of reference for Weight is W and name of reference for ch is T. In the program the values of the
three references are changed. From the output we find that values of the variables are also changed
to the values of the corresponding references. Also note that the address of n and Count is same,
i.e. 0012FF7C. Also the variable Weight and its reference W have the same address where the
value is stored.

PROGRAM 9.19 – Illustrates declaration of references to variables.

#include <iostream>
using namespace std;
int main()

{int n = 100 ;
double Weight = 35.6;
char ch = ‘S’;

int& Count = n ; // Count is a reference to n.

double& W =Weight; // W is a reference to weight.
char& T = ch; // T is reference to ch.
cout<<“n = ”<<n<<” \t&n = ” <<&n <<“\n”;
cout<<“Count = ”<<Count<<“ \t &Count = ” <<&Count << endl;

cout<< “Weight = ” <<Weight << “, &Weight = ”<< &Weight<<“\n”;

cout<< “W = ” << W<< “ \t &W = ”<< &W<<“\n\n”;

cout<<“ch = ”<<ch <<“\t T = ” << T <<endl;
T = ‘Z’;
n = 50; //The value of reference n is changed to 50
W = 12.5; // The value of reference W is changed to 12.5

cout<< “n = ”<<n << “ \tCount =”<< Count <<endl;

cout << “W = ”<< W<< “ \tWeight =” << Weight << endl;

cout << “ch = ” << ch <<“ \t T = ” << T << endl;

return 0;
}

The output is given below.

n = 100 &n = 0012FF7C
Count = 100 &Count = 0012FF7C
❖ 220 ❖ Programming with C++
Weight = 35.6, &Weight = 0012FF74
W = 35.6 &W = 0012FF74
ch = S T=S
n = 50 Count =50
W = 12.5 Weight =12.5
ch = Z T=Z
The output of the program shows that the address of the variable n and its reference Count
is the same. If the value of reference Count is changed from 100 to 50, the value of n is also
changed to 50. So dealing with a reference is as good as dealing with the variable. Similar results
are obtained with double. Weight and the char ch.
In the following program it is shown that any number of references may be created and
dealing with any of the references is as good as dealing with variable. A reference may be defined
from another reference or a pointer as well.

PROGRAM 9.20 – Illustrates multiple references to a variable.

#include <iostream>
using namespace std;
int main()
{ int n = 60 ; // variable n = 60
int* ptrn = &n; // ptrn is pointer to n
int& Count = *ptrn ; // Count- a reference to n created from
// pointer to n
int &m = n; // m is another reference to n
int& T = Count; // T is reference created by a reference
cout<<“Count = ”<<Count<<“\t&Count = ” <<&Count<<endl;
cout<<“*ptrn = ”<<*ptrn<<“ , \tptrn = ” <<ptrn <<“\n”;
cout<<“T = ”<<T<<“ , \t&T = ”<<&T<<endl;
cout<<“n = ”<<n<<“ , \t&n = ” <<&n <<“\n”;
cout<<“m = ”<<m<<“ , \t&m = ” <<&m <<“\n”;
T = 80; // value of T changed to 80
cout << “T = ”<<T<<“ \t*ptrn = ”<<*ptrn<<endl;
cout<< “n = ”<<n << “, Count =”<< Count << “, m = ” << m <<endl;
return 0;
}

The expected output is given below.

Count = 60 &Count = 0012FF7C
*ptrn = 60 , ptrn = 0012FF7C
T = 60 , &T = 0012FF7C
 Pointers ❖ 221❖
n = 60 , &n = 0012FF7C
m = 60 , &m = 0012FF7C
T = 80 *ptrn = 80
n = 80, Count =80, m = 80
In the above program Count, m, T are references to n. Count is created by the pointer to n
while T is created by another reference Count. The reference m is created directly by n. In all
these cases, the references behave in the same manner. All of them have the same address, i.e.
0012FF7C. If the value of any of the reference is changed, the value of all references including
the variable changes. Manipulating any one reference is as good as manipulating the variable.

9.13 PASSING ARGUMENTS BY VALUE AND BY REFERENCE

PASSING ARGUMENTS BY VALUE

The arguments to a function may be passed on by values or by references. This topic has already
been dealt briefly in Chapter 7 on Functions. When the arguments are passed on by values, the
copies of the values of the variables are passed on to the function. The function may operate on
these copies, i.e. may change their values, but the function does not get to the variables and so
it cannot change their values. The benefit of this is that the function cannot change the original
data. However, when the values are passed on by reference or by pointers, in fact, we pass on
the variables to the function. Now if the function changes their values, the changed values get
stored in the memory blocks allocated to variables. So the values of variables are also changed.
It is explained by an example below. Let us define a function F() which has two integer parameters
a and b as given below.
int F ( int a, int b) // Function definition
{ a +=3;
b +=2;
return a*a + b*b; }
int main()
{ int P = 3, Q =2;
Z = F(P,Q);
cout<<Z << endl; }

In the above code a function with parameters int a and int b changes the values of a
and b in its definition body and then function is evaluated taking changed values of a and b. In
the main() when this function is called as F(P, Q) the copies of values of two integer variable P and
Q are passed to it. The variable a takes the value of P and b that of Q. So a becomes 3 and b
becomes 2. But variables themselves, i.e. P and Q are not passed on to the function, only their
copies are passed to the function. The function in fact does not know where the variables P and
Q are stored. The function as defined above changes the values of these copies which are passed
on to it. The values of a and b are changed in the function to 6 and 4 respectively and function is
evaluated. But values of variables P and Q do not change. This is called passing arguments by value.
Passing the arguments by value has the advantage that the original data is not changed,
❖ 222 ❖ Programming with C++

but the disadvantage is that when data is very large the process would occupy more memory space
as well as the execution of program will take more time. In some functions it is desired to change
the values of variables, such as in swapping of values, in such cases we will have to pass on the
arguments by reference.
Program 9.20 is an illustration of passing arguments by values. In the program a function
Sum() has two integer parameters a and b. When the function is called in the main () the values
of P and Q are passed on to function as a and b respectively. The values of a and b are changed
in the function Sum () as explained above but the values of P and Q are not changed at all.

PROGRAM 9.21 – Illustrates passing of arguments by values.

#include<iostream>
using namespace std;
int Sum(int a, int b )
{a +=3;
b+=2;
return a*a + b*b ;
}
int main()

{ int P,Q,Z ;
P = 3;
Q= 2;
cout<<“Before calling function the values are: ”<<endl;
cout<< “P = ” <<P<<“, Q = ” <<Q<< endl;
Z = Sum(P,Q);

cout<<“After the first call the values are :”<<endl;

cout << “ Z = ” <<Z<<“, P = ” << P<<“, Q = ” <<Q <<endl;
cout<<“After the 2nd call the values are:”<<endl;
Z = Sum (P,Q);
cout << “Z = ” <<Z<<“, P = ” << P<<“, Q = ” <<Q <<endl;
return 0;
}

The expected output is given below.

Before calling function the values are:
P = 3, Q = 2
After the first call the values are :
Z = 52, P = 3, Q = 2
After the 2nd call the values are:
Z = 52, P = 3, Q = 2
 Pointers ❖ 223❖

PASSING ARGUMENTS BY REFERENCE

Now, for passing the variable by reference the same function is defined as given below.
int F( int &a, int &b)
{ a +=3;
b +=2;
return a*a + b*b;
}

In such a case a and b are references to P and Q. Since the values of a and b are changed in
the body of the function the values of P and Q will also get changed. In Program 9.22 given
below, the variables are passed to the function through references. So change in the value of
references will change the value of variables also. Thus we see from the output of following
program that values of P and Q after the call of the function are changed. See carefully the
method of defining the function. In the following listing the function has been called twice. Each
time a different result is obtained, because the values of P and Q get changed in every call of
function.

PROGRAM 9.22 – Illustrates passing arguments by reference.

#include<iostream>
using namespace std;
int Sum(int &a, int &b) //&a and &b are references of variables.
{a +=3;
b+=2;
return a*a + b*b ;
}
int main()
{
int P,Q,Z ;
P = 3;
Q= 2;
cout<<“Before calling function.”<<endl;
cout<< “P = ” <<P<<“, Q = ” <<Q<< endl;
Z = Sum(P,Q); // first call of function
cout<<“After the first function call”<<endl;
cout << “Z = ” <<Z<<“, P = ” << P<<“, Q = ” <<Q <<endl;
Z = Sum(P,Q); // second call of function
cout<<“After the second function call”<<endl;
cout << “Z = ” <<Z<<“, P = ” << P<<“, Q = ” <<Q <<endl;
return 0;
}
❖ 224 ❖ Programming with C++

The expected output of the program is as under.

Before calling function.
P = 3, Q = 2
After the first function call
Z = 52, P = 6, Q = 4
After the second function call
Z = 117, P = 9, Q = 6
The output of the above program shows that before the function call the values of variables
are P =3 and Q =2. After the first call P becomes 6 and Q becomes 4. The value of the function
Sum (P, Q) is = 52. After the second call of the function the values of P and Q are changed again.
P is now equal to 9 and Q is equal to 6. The value of Sum () now becomes 117.

9.14 PASSING ARGUMENTS THROUGH POINTERS

Pointers keep the actual address of the variables, therefore, passing arguments through pointers
is as good as passing on the variables to the function. Program 9.23 illustrates this fact.

SWAP( ) FUNCTION
Here we define Swap function which exchanges the values of two int variables (this Swap ()
function is different from template function swap () of C++ Standard library, because S is in
upper case. The function is defined so that values are passed on by pointers. The values are
exchanged as expected.

PROGRAM 9.23 – An illustration of passing arguments by pointers.

#include <iostream>
using namespace std;
void Swap(int *px, int*py) // Definition of Swap function
{int temp;
temp = *px;
*px = *py;
*py = temp; }
int main()
{
int m ,n ; // Declaration of two integers.
cout<<“Enter two integers : m = ”; cin>> m ; cout<< “n = ” ;
cin >>n; // values of m and n are entered by user.
Swap (&n , &m );
cout <<“After swapping the values are :”;
cout<<“ m =”<<m<< “, n = ” << n<<endl;
return 0;
}
 Pointers ❖ 225❖

The expected output is given below.

Enter two integers : m = 50
n = 120
After swapping the values are : m =120, n = 50

SORTING AN ARRAY WITH POINTERS

The Swap () function as defined above has been used to sort an array of integers in descending
order. See the following program for illustration.

PROGRAM 9.24 – Illustrates sorting of array using pointers.

#include <iostream>
using namespace std;

void Swap(int *px, int *py) // Declaration and

{ int temp ; // definition of Swap() we are
temp = *px; // not using swap() of C++ std Lib.
*px = *py;
*py = temp;}

int main()
{
int kim [ ] = { 60,50,86,90,120,67,89 } ;

int* pkim = kim; /* declaration and initialization of pointer to array kim[]*/

for ( int j = 0; j<=6;j++)

for ( int k =0; k<6–j ; k++)
if (*(pkim +k)< *(pkim +k+1))
Swap((pkim+k), (pkim + k+1));

for ( int i=0 ; i<=6;i++)

cout << *(pkim +i) <<“ ”;
return 0;
}

The sorted list is given in the output as under.

120 90 89 86 67 60 50
Application of pointers in selecting greater of two numbers
The following program is another example of using pointers for selecting greater of the two
numbers.
❖ 226 ❖ Programming with C++

PROGRAM 9.25 – An application of pointers in finding greater of two numbers

#include <iostream>
using namespace std;
int Max(int *px, int*py) // definition of Max()
{ return (*px>*py ? *px:*py); }
int main()
{
int m ,n ;
cout <<“Enter two integers :”; cin>> m>>n;
int Maxnum = Max (&n,&m);
cout<<“Maximum of the two numbers is “ << Maxnum<<endl;
return 0;
}

The expected output is as below.

Enter two integers :65 32
Maximum of the two numbers is 65

9.15 POINTER ARITHMETIC

It has already been illustrated in Program 9.2 that a pointer may be assigned if the type of both
is same. The assignment would result in having two pointers pointing to same value stored in
memory. The other operators that may be applies to pointers are +, – and increment (++) and
decrement (– –) operators
The relational operators, in general, have no meaning for pointers pointing to unrelated
objects. However, they may be used when dealing with elements of same array, in which case,
the pointer values (addresses) of different elements may be compared. The pointers may be
incremented or decremented to access different elements of the array. The pointer increment,
however, increases the value of pointer by the number of bytes occupied by one element of array.
Thus if we are dealing with an array of integers, each element occupies 4 bytes. Let us declare a
pointer as below.
int A[5] = { 6,8,9,8,4};
int * ptrA = A ;
In the above declaration ptrA points to the first element of array i.e. A[0]. Now ++ptrA
points to the next element of array. Similarly (ptrA + 3) will point to A[3], i.e. the 4th element
of array. The actual value of pointer increases by 4*3 = 12 bytes. The value of array element may
be obtained by dereference operator. Thus *(ptrA + 3) gives the value of A[3]. Please note that
the expression *(ptrA + 3) should not be written as *ptrA+3, which will be equivalent to addition
of 3 to value of first element (A[0]) of array.
 Pointers ❖ 227❖

9.16 VOID POINTERS

As explained above, the pointers may be assigned if both are of same type. Otherwise, we need
to convert the pointer on the right side to the type of pointer on the left side of equality sign
(=). The void pointer ( void * ) can, however, represent any pointer type. Thus any type of pointer
may be assigned to a void pointer. But the reverse is not valid. A void pointer cannot be assigned
to any other type of pointer without first converting the void pointer to that type.
Also the void pointer cannot be directly dereferenced, because it is pointer without a type.
The compiler does not know the number of bytes occupied by the variable to which the void
pointer is pointing. It is illustrated in the following program.
PROGRAM 9.26 – Illustrates the application of void pointers.

#include<iostream>
using namespace std ;

int main()
{
int x = 8, *px; // declaration of int and pointer for int
double D = 5.8, * pd; // declaration of double and
// pointer for double
px = &x;
pd = &D;
cout<<“x = ” << *px<<endl;
cout <<“D = ” << *pd<<endl;
void* pv ; // declaring a void pointer
pv = &x ; // assigning address of int to void pointer
cout<< “Now getting values through void pointers”<<endl;
cout <<“x =” << *(int*)pv << endl ;
// pv is cast to type int for dereferencing

pv = &D; // assigning address of double to void pointer

cout<<“D = ” << *(double*)pv<< endl;
// pv is cast to type double for dereferencing.
return 0 ;
}

The output is given below.

x=8
D = 5.8
Now getting values through void pointers
x =8
D = 5.8
❖ 228 ❖ Programming with C++

9.17 SUMMARY OF POINTER DECLARATIONS

Table 9.1 – Summary of useful declarations in pointers

Declaration Description Example

int *ptr; Pointer to an integer. int y ;
int *ptry = &y;
double *ptrd; Pointer to a double. double* ptrPI = &PI;
char *ptrc; Pointer to a character. char ch = ‘A’;
char*ptrc= & ch;
void* ptr; Pointer to void. void *memset(void *S,
int c, size_t n)
const int *ptr; Pointer to a const integer. const int n ;
const int* ptrn = &n;
int* const ptr ; Constant pointer to an int n;
integer. The pointer int* const ptr = &n;
cannot be changed.
const int* const ptr ; Constant pointer to a const int n ;
constant integer. const int*const ptr
Both cannot be changed. = &n ;
int ** pptrn ; Pointer to pointer to int n ;
an integer. int* ptrn = &n;
int**pptrn= &ptrn;
int& Count = n ; Count is reference to n. int& Count = n ;
int A[4];int* ptrA = A ; Pointer to single dimensional int A[4];
int array A. int* ptrA = A;
int (*ptr)[]; Pointer to two dimensional int B[3][4];
array. int (*ptrb)[4]= B;
int *A[6]; Array of 6 pointers to integers. int *A[6];
int(*Function)(int,int); Pointer to function returning int(*Funct)(int, int);
an int value and with two
int parameters.
double(*Function[])(int, An array of pointers to double(*Funct[])
int ); functions returning double (int,int);
and having two int parameters.
void (*function)(void); Pointer to a void function void(*function) (void);
with void arguments.
 Pointers ❖ 229❖

EXERCISES
1. What is a pointer? How is it declared for an integer?
2. An integer variable x has the value 50. Write the code for declaration of the variable, its
reference by name myx and pointer by name ptrx.
3. Write a test program to declare a float variable y, initialize it to zero. Create a reference by
name refy. Now change y to 3.675. Find the value of refy and its memory location.
4. What are the sizes (in bytes) of memory spaces allocated for pointers to (i) a character variable,
(ii) a long int, (iii) a float number, (iv) double number.
5. Write code for declaring pointer to (i) one dimensional array, (ii) a two dimensional array.
6. How do you get the value of variable from its pointer to pointer?
7. Write a program to declare an int variable and its pointer and to determine its square and cube
using its pointer.
8. How do you declare pointer to pointer for (i) int variable, (ii) for double variable.
9. What is wrong with the following codes.
(i) int* ptrn = n;
(ii) double ptrnPI = 3.14159;
(iii) char * ch = A ;
10. What is difference between a reference and a pointer?
11. Declare a reference with name ref_grade for character grades like A, B, etc.
12. What do you understand by the following.
(i) int (*Function)(int, int) ;
(ii) double(*Function[])(int, int, double)
13. Explain the difference between following two declarations.
(i) int (*A)[6];
(ii) int *A[6];
14. What is the difference between following declarations?
(i) const int*ptr ;
(ii) int * const ptr;
(iii) const int * const ptr;
15. PA is a pointer to an array A[10]. Which of the following codes are correct to get the next
element of the array?
(i) *PA +1;
(ii) *(PA+1) ;
(iii) *A++
16. Write a program for illustrating the use of pointers for traversing an array.
17. Write a test program to illustrate the application of oprators new and delete for int variables
and for character variable.
❖ 230 ❖ Programming with C++
18. When new[] is used to declare an array of integers, how do you initialize the elements and
write code for their output and their deletion.
19. Write code of a program for preparing lists of names by user of program wherein the total
number of entries are unknown. Make use of pointers.
20. Write a program to illustrate the application of new[] and delete[]

❍❍❍
CHAPTER

10.1 DECLARATION OF A C-STRING

A C-string is a sequence of characters stored in sequential memory blocks like an array and it is
terminated by Null character (‘\0’). Unlike arrays, a string is treated as a single unit. String
characters may include letters, digits and special symbols such as +, –,*, [],(), etc., but not the
escape characters. The strings based on class <string> are called C++ strings and are dealt separately
in Chapter 17. The C++ Standard Library has many useful functions for manipulation of C-
strings as well. Here in this chapter we shall deal with C-strings, however, we shall refer to them
as strings. A C-string may be declared and initialized as illustrated below.
char Name [8] = “John”;
Here Name is the identifier (name) of string . The names of strings are decided as we do
for any other variable. The number 8 in square bracket indicates that there could be 8 characters
in the string including the Null character (‘\0’) which is appended by the system to mark the
end of string. This string is initialized as John. In assignment like this it is required to enclose
the sequence of characters between double quotes as “John”. When a string is initialized with
a string of characters in double quotes the system appends the null character at the end of string.
In this case system would allocate 8 memory blocks (each of size one byte). Since the string
Name[8] is also an array the different elements of this array are as below.
Name[0] = ‘J’;
Name[1] = ‘o’;
Name[2] = ‘h’;
Name[3] = ‘n’;
Name[4] = ‘\0’;
The remaining three memory blocks are empty. The last character Null (‘\0’) marks the
end of string. Thus if we determine the size of Name we would get 8 because it has been declared
so. However, if the string is declared and initialised as
char Name[] = “John”;
we would get the size as 5, i.e. 4 characters of John and one Null character (‘\0’).
❖ 232 ❖ Programming with C++
PROGRAM 10.1 – Illustrates declaration and initialization of C-strings.

#include <iostream>
using namespace std;
int main()
{
char Name1[8] = “John”;
char Name2[] = “John”;
char Symbols[] = “@ ! # $ % 5 , ~ ^ & * () _ + = \ ;: {} () [] \t\n\a”;
cout<<Name1<<endl ;
cout<<Name2<<endl;
cout<<sizeof(Name1)<<endl;
cout<<sizeof(Name2)<<endl;
cout<<“Name1 is “<<Name1<<endl;
cout<<“Name2 is “ << Name2<< endl;
cout<< “Symbols are “<< Symbols<< endl;
for(int i =0; i<8;i++)
cout<<Name1[i] <<“ “;
cout<<endl;
return 0 ;
}

The expected output is given below.

John
John
8
5
Name1 is John
Name2 is John
Symbols are @ ! # $ % 5 , ~ ^ & * () _ + = ;: {} () []
John

From the output statements and the output it is clear that the strings Name1, Name2, and
Symbols have been treated as single units. The size of Name2 is 5 because it also includes the
null character besides the 4 characters in John. Many symbols have been put in the string Symbol
simply to show that these could also be part of string. You would notice that the escape
sequences are not part of strings. These are not there in the 7th line of output. It should be
noted that the characters in string Symbols are only characters, these do not have the usual
meaning. For example, here in string Symbol the character ‘5’ does not have the usual value
five but is equal to 53 as per ASCII code (see Appendix A). The output of Name1, character
by character shows only 4 characters of John, the remaining 4 characters are not shown in output.
Out of these one is ‘\0’ character which does not show up in output and the other 3 memory
 C-Strings ❖ 233❖
blocks are empty. In the computer memory string characters are stored in adjacent memory
blocks of one byte each as an array of characters terminated by a Null character (‘\0’) as illustrated
in Fig. 10.1.
Name1
0 1 2 3 4 5 6 7 array subscripts

J o h n \0

Address – 0012FF78 byte number of first byte in memory

Fig.10.1: The string Name1[8] = “John”; is stored as an array in computer.

The shaded bytes are empty. Array subscripts are shown above the blocks

The following program illustrates some differences between the normal arrays and character
arrays and strings.

PROGRAM 10.2 – Illustrates differences between arrays and strings.

#include <iostream>
using namespace std;

int main()
{
int Array [] = { 5 , 6 , 7 , 8 }; // int array
char B[] = { ‘U’, ‘V’, ‘X’,’Y’}; // char array

char Name[] = “John”; // string

char* S = “You are Welcome to C++!” ; // string pointed to by S

cout<<“Name = ”<<Name<<endl;
cout<<“*Name = ” << *Name<<endl;
cout<<“S = ”<<S<<endl;
cout<<“*S = ”<<*S<<endl;
cout << “Array = ” <<Array<<endl;
cout << “*Array = ”<< *Array<<endl;
cout<<“B = ”<< B<<endl;

cout<<“*B = ” <<*B<<endl;
return 0 ;
}
❖ 234 ❖ Programming with C++

The expected output is given below.

Name = John
*Name = J
S = You are Welcome to C++!
*S = Y
Array = 0012FF70
*Array = 5
B = UVXY?
*B = U

The array Array is an int array. When its name is called for output it gives the address of
first element of the array. And when *Array is called for output it gives the value of the first
element of the array. These are usual with arrays. Similarly when *Name is called for output it
gives the first element of string Name that is J. So is the case with *S and *B. The output of
*S is the Y the first character of string pointed by S and *B gives U the first element in the char
array. However, when name of character array or name of string is called for output instead of
giving address of first element it gives the complete string as output. Similarly when pointer S
to a string is called for output it gives the full string as output while in other arrays it would
give the address.

10.2 INPUT/OUTPUT OF C-STRINGS

The object cin of <istream> along with extraction operator >> may be used for input of strings
if there is no white space between the characters because cin takes the white space as the end of
string. However, there are other functions like cin.get() and cin.getline() which may
also be used for input of strings. These functions read white spaces as well. The cin.get()
reads character by character while cin.getline() reads the whole line.
In the following program only cin is used. The input string consists of two words Mona
Lisa, but cin extracts only the first word Mona and stops reading on encountering white space.
PROGRAM 10.3 – Illustrates string input by cin.

#include <iostream>
using namespace std;
int main()
{
char S[15] ;
cout<<“Write a short name : ” ;

cin >> S ;
cout <<“You have written : ”<< S<<endl;
return 0 ;
}
 C-Strings ❖ 235❖
The expected output is below.
Write a short name : Mona Lisa
You have written : Mona
From the output it is clear that cin stopped reading after encountering the white space, i.e.
after Mona. Now if the same string is extracted by cin character by character as an array, the
complete name is read but the white space is not read as space. This is illustrated in the following
program. The two words Mona and Lisa are merged together.

PROGRAM 10.4 – Illustrates limitations of cin in reading string.

#include <iostream>
using namespace std;
int main()
{ char S[15];

cout<<“Write a short name : ”;

for (int i =0; i<=15;i++) // Use of for loop for reading

{ cin>>S[i] ; // It does not read the white space
cout<<S[i] ;
}
return 0 ;
}

The expected output is given below.

Write a short name : Mona Lisa
MonaLisa

From the 2nd line of the output it is clear that the white space has not been read and included
in the string. So in the output the two parts of name are merged together.
FUNCTION CIN. GET()

The above mentioned problem is not there if we use function cin.get() and function cin.getline
(). The use of cin.get() is illustrated below. It has three arguments.
cin.get(identifier, n, ch);
Here identifier is the name of string, the function reads a maximum of n–1 characters and
ch is the name of character on encountering which the reading would stop. The following three
programs illustrate the application of cin.get() function with different number of arguments.
Below in Program 10.6 the function call is coded as,
cin.get (S, 15);
Here S is the name of the string, n = 15. This program limits the number of characters for
input by cin.get() to 14 even though the string may have more characters (the 15th character is
❖ 236 ❖ Programming with C++
‘\0’ character). If the input is user interactive as in online filling of forms etc., the number n
should be sufficiently large so that user input data is not lost.

PROGRAM 10.5 – Illustrates cin.get() function for reading strings up to specified number
of characters.

#include <iostream>
using namespace std;
void main()
{char S[15];
cout<<“Enter a name :” ;
cin.get(S,15) ;
cout <<“you have written :”<< S<<endl;
}

The output given below shows that full name along with white space has been read. Compare
this output with that of the last program in which only cin has been used.
Enter a name :Mona Lisa
you have written :Mona Lisa

PROGRAM 10.6 – Illustrates action of cin.get () when specified number is too small.

#include <iostream>
using namespace std;
void main()
{char S[15];
cout<<“Enter a name :” ;
cin.get(S,5) ;
cout <<“you have written :”<< S<<endl;
}

The output given below shows that only the first part of name, i.e. Mona has been read
because it was specified to read only five characters in the statement cin.get (S, 5);.
Enter a name :Mona Lisa
you have written :Mona
As mentioned above, we can as well specify a character in cin.get()on encountering of
which the reading would stop. In the following program we specify a character ‘s’ in the function
cin.get () on encountering of which cin.get() stops reading.

PROGRAM 10.7 – Illustrates action of cin.get() with a delimiting character.

#include <iostream>
using namespace std;
void main()
 C-Strings ❖ 237❖
{ char ch = ‘s’;
int i =0;
char B[20];
cout<<“Enter a name : ” ;
cin.get(B,20,ch);

cout <<“You have written : ”<< B <<endl;

}

The expected output is given below.

Enter a name : Johnson
You have written : John
The above output shows that cin.get() has read only up to John. On encountering
character ‘s’ it has stopped reading. The delimiting character ‘s’ has not been taken in.
The function cin.get() reads the string character by character. One benefit of reading
character by character is that you can also operate on any particular character. You may count
the number of characters or count the number of occurrences of a particular character in a string
or replace a particular character with another character while the characters are read. The following
program counts the total number of characters in a statement and the number of times ‘o’ occurs
in a statement.
PROGRAM 10.8 – Illustrates that while reading with cin.get() the characters read may be
counted.

#include <iostream>
using namespace std;
int main()
{
int n =0, m = 0;
char para;

cout<<“Enter a few words : \n” ;

while (cin.get(para))
{if ( para !=’ ‘)
++ n ;
if ( para == ‘o’)
++m;
if ( para == ‘.’)
break; }

cout <<“n = ”<<n <<“, m = ”<<m <<endl;

return 0;
}
❖ 238 ❖ Programming with C++

The expected output is as below.

Enter a few words :
Mohan and Sohan are going to market.
n = 30, m = 4
The output of the program shows that the total number of characters are 30 and ‘o’ occurs
four times. The white spaces are not counted. In the above program the while loop is not limited
by white space so program does not come to an end. It is an endless loop. So we have put the
statement break; to exit from the loop on encountering full stop.
Function cin.getline ()
The function cin.getline() reads all the characters in a line until the limiting number
of characters is reached or it encounters a delimiting character. It takes maximum of three
arguments. The syntax for function call is as below.
cin.getline( identifier, size_t n, ch);
The first argument in the bracket is the name of string we want to put in. The integer n is
used to limit the number of characters to be read. The function stops reading when (i) n–1
characters have been read or (ii) a new line or end of file is encountered or (iii) the delimiting
character ch is encountered. It will not produce an error even if ch is not specified because n ,
end of line or EOF can do the same job. In the following, Programs 10.9, 10.10 and 10.11
illustrate the applications of this function with different number of arguments.

PROGRAM 10.9 – Illustrates use of cin.getline() to read input from keyboard.

#include <iostream>
using namespace std;
int main()
{
char B[50];
cout <<“Enter a long name: “;
cin.getline(B,50) ; // It specifies to read up to read 50–1
// characters from string. The last character is Null (\0).
cout<<“You have written = “<<B <<“\n” ;
return 0 ;
}

The expected output is given below.

Enter a long name: Mohan Dass Karamchand Gandhi
You have written = Mohan Dass Karamchand Ghandhi

PROGRAM 10.10 – Illustrates application of cin.getline() function.

#include <iostream>
using namespace std;
 C-Strings ❖ 239❖
int main()
{
char B[20];
cout <<“Enter a name: ”;
cin.getline(B,5) ; // read first 5 characters of B

cout<<“You have written = ”<<B <<“\n” ;

return 0 ;
}

The name entered was Mona Lisa but the function has read only first five characters, i.e. Mona
as given in the output given below, because, it was specified so in getline() function.
Enter a name: Mona Lisa
You have written = Mona
In the following program a delimiting character has been put in the cin.getline() function to
stop reading on encountering the character.

PROGRAM 10.11 – Illustrates use of cin.getline() to read up to specified character.

#include <iostream>
using namespace std;
int main()
{
char B[20];

cout <<“Enter a name: ”;

char ch = ‘s’;
cin.getline(B,20, ch) ; //read up to character ‘S’
cout<<“You have written = ”<<B <<“\n” ;
return 0 ;
}

The output is given below. It stopped reading on occurrence of letter ‘s’. So only Mona Li is
read as given in the output below.
Enter a name: Mona Lisa
You have written = Mona Li

10.3 STANDARD FUNCTIONS FOR MANIPULATING STRING ELEMENTS

Function cin.putback( )
With the help of this function a character may be replaced by another character in a string.
See the following program in which the if (expression) examines if the character is ‘M’ and the
statement cin.putback(‘S’); replaces it with character ‘S’.
❖ 240 ❖ Programming with C++
PROGRAM 10.12 – Illustrates the action of function putback().

#include <iostream>
using namespace std;
void main()
{
char ch;
cout<<“Enter a sentence : ” ;
while (cin.get(ch))
{ if (ch ==’M’)
cin.putback(‘S’);
else
cout<<ch;
}}

The expected output is given below. ‘M’ has been replaced by ‘S’.
Enter a sentence : Mona and Madu went to Madras.
Sona and Sadu went to Sadras.
Functions cin.peek() and cin.ignore ()
The function cin.peek() takes note of a character, i.e. if the desired character is encountered
it will give a signal but the function itself does not do anything, however, the signal may be
used by other functions to carry out the desired process. In the following program cin.peek()
has been used to notice the character ‘n’ while cin.ignore() function has been used to ignore it.
The function call for cin.ignore() is coded as below.
cin.ignore(n, ch) ;
Here n is the maximum number of characters to ignore and ch is the termination character.

PROGRAM 10.13 – The program illustrates functions cin.peek(), cin.ignore().

#include<iostream>
using namespace std;
int main()
{ int m =0 ;
int k = 0;
char ch ;
cout<<“Enter a sentence :\n” ;
while (cin.get(ch)) // unending loop
{ ++k;
if (ch ==’M’)
{++m;
cin.putback(‘S’);}
else
 C-Strings ❖ 241❖
cout<<ch;
while (cin.peek()==’n’)
cin.ignore(1,’n’);
if (ch == ‘.’) break;} // break statement
cout << “\nTotal number of characters read = ” << k <<endl;
cout<<“Number of M are = ” << m <<“\n” ;
return 0;
}

The expected output is given below.

Enter a sentence :
Madhu , Malti and Mamta all will come on Monday.
Sadhu , Salti ad Samta all will come o Soday.
Total number of characters read = 49
Number of M are = 4

In the above program, in the first while loop, it finds if the character is ‘M’, if so, it replaces
it with ‘S’. In the second while loop the function peek() notices for the character ‘n’ and the
function cin.ignore() ignores it. The variable k counts the total number of characters read while
m counts the number of M occurring in the sentence.

10.4 CONVERSION OF C-STRING CHARACTERS INTO OTHER TYPES

FUNCTIONS ATOI(), ATOL(), ATOF()

In some programs strings may contain digits or digits with a decimal point etc. In a string the
digits are stored as characters and not as the values that the digits represent. The functions
atoi(), atol(), atof() of the general utilities header file <cstdlib> convert the character
digits into digital values. The individual details of these functions are given below.
atoi() It converts its argument consisting of a string of digits into int type number
and returns int value. However, if the first character is not a digit the string
may not be converted, in that case, it returns 0. The syntax is given below. In
this the chptr is pointer to character constant.
int atoi ( const char *chptr)
atol() It converts its argument consisting of a string of digit characters into long type
number and returns the number. However, if the first character is not a digit
the string may not be converted and in that case it returns 0.The syntax is,
int atol ( const char *chptr)
atof() It converts string of digit characters with decimal point representing a floating
point number into double type number and returns the double number.
However, if the first character is not a digit the string may not be converted
and in that case it returns 0.The syntax is given below.
int atof ( const char *chptr)
❖ 242 ❖ Programming with C++

The following program illustrates application of the three functions.

PROGRAM 10.14 – Illustrates use of atoi(), atol() and atof() on strings.

#include <iostream>
# include <cstdlib>
using namespace std;
void main()
{
char ch1 []= “6444”;
char ch2[] = “21.66”;
char ch3[] = “Absent45”;
char ch4[] = “45Absent”;
char chL1[] = “Delhi45676548”;
char chL2[] = “45676548Delhi”;
cout<<“Number represented by ch1 divided by 2 = ” << atoi(ch1)/2 <<endl;
cout << “Number ch2 /3= ” << atof (ch2)/3<<endl;
cout << “Number ch3 = ” << atoi (ch3)<<endl;
cout << “Number represented by ch4 = ” << atoi (ch4)<<endl;
cout << “Number represented by chL1 = ” << atol (chL1)<<endl;
cout << “Number represented by chL2 = ” << atol (chL2)<<endl;
}

The expected output is as below.

Number represented by ch1 divided by 2 = 3222
Number ch2 /3= 7.22
Number ch3 = 0
Number represented by ch4 = 45
Number represented by chL1 = 0
Number represented by chL2 = 45676548
Six strings ch1, ch2, ch3, ch4, chL1 and chL2 are declared and initialized. The functions
atoi(), atof () and atol() are called to convert the strings into corresponding digits. The strings
which begin with non digit characters are not converted and 0 is returned while those which
begin with digit characters are converted. To show that the converted digits represent number,
the numbers obtained in first two cases have been divided by 2 and 3 respectively. The trailing
ends of strings consisting of non-digital characters are ignored.

FUNCTIONS strtod (), strtol () and strtoul()

In case of functions atoi(),atol() and atof()we saw that if the digits are followed by
non-digital characters, the non-digital characters are ignored. This is not the case with the
functions strtod(), strtol() and strtoul()which are explained below.
strtod() The function converts the digital part of string to double number and returns the
double value. The second part is extracted by the pointer to the substring. The
 C-Strings ❖ 243❖

function arguments comprise pointer to string and address of pointer to substring.

The syntax is given below.
double strod(const char* chptr, char** subptr)
strtol() The function strtol() takes three arguments, the first is the string name/pointer to
string, the second is the address of the pointer to non-digital substring and the third
is an integer which specifies the base of the value being converted. The code is given
below.
long strol(const char* chptr, char** subptr, int base)
strtoul() The function converts the digits and returns unsigned long number. The function
strtoul()also takes three arguments, the first is the pointer to string, the second is the
address of the pointer to the non-digital sub-string and the third is the base (8 for
octa, 10 for decimal and 16 for hexadecimal) of the value being converted. The code
is as below.
unsigned long stroul(const char* chptr, char** subptr, int base)

The application of strtol() is illustrated in the following program.

PROGRAM 10.15 – Illustrates functions strtol().

#include <iostream>
# include <cstdlib>
using namespace std;
void main()
{ long n ;
int m ;
char ch1 []= “100 Delhi 110016 ”;
char ch3[] = “D64”;
char *psubstring1;
char *psubstring2;
m = strtol(ch3, &psubstring2,10);
n = strtol(ch1, &psubstring1, 10);
cout<<“Converted digital (base 10) = n = ” <<n<<endl;
n = strtol(ch1, &psubstring1,16);
cout <<“n ( base 16) = ”<<n<<endl;
cout<<“Original string = ” <<ch1<<endl; // the original string

cout<<“The remainder substring = ”<<psubstring1<<endl<<endl;

cout<<“Original string = ” <<ch3<<endl;
cout<<“Converted string part = m = ” <<m<<endl;
cout<<“The remainder substring = ”<<psubstring2<<endl;
}
❖ 244 ❖ Programming with C++

The expected output is given below.

Converted digital (base)10 = n = 100
n ( with base16) = 256
Original string = 100 Delhi 110016
The remainder substring = Delhi 110016
Original string = D64
Converted string part = m = 0
The remainder substring = D64
In the above program two strings ch1 and ch3 are declared. String ch1 starts with digits
while ch3 starts with letter D. In case of ch1, in the first conversion the base is taken 10 so 100
gets converted into 10 × 10 = 100, in the second conversion the base is 16 (hexadecimal) so in
this case 100 gets converted to 16 × 16 = 256 that is square of 16. Similarly, other numbers
may be taken as base. It would be better to consult the compiler document. The substring
Delhi 110016 is also recovered. From output it is clear that ch3 has not been converted and 0
is returned because it does not start with a digit. The complete string is recovered as unconverted
sub-string (D64).

10.5 ARRAYS OF C-STRINGS

If it is desired to make a list of statements such as list of names, list of quotations, list of books,
etc., we would be dealing with arrays of strings. An array of strings is a two dimensional array
of type char and may be declared as illustrated below.
char identifier [n][m];
Here identifier is the name of the array, n is the number of strings and m is the number of
characters in each string including the null character. The value of m may be decided so that a
string having the maximum characters is covered. The Fig. 10.2a shows an array of strings in
which a list of four names is displayed. Here n = 4 and m = 9. In this arrangement each string
is allocated 9 bytes and can have 9–1 characters of string and the terminating null character. If
a string has fewer characters than 8 characters the remaining memory blocks remain empty, and
is wasteful for memory space. A better method of declaration is
char* identifier [n]
It is an array of pointers of type char. It is illustrated in the following program.

PROGRAM 10.16 – Illustrates an array of strings.

#include <iostream>
using namespace std;
int main()
{
char Name [4][9] = {“Sun”, “OurEarth”, “Sky”, “Mars”};
cout <<“Size of Name = ” << sizeof(Name)<< endl;
char * name[4] = {“Sun”, “OurEarth”, “Sky”, “Mars”} ;
 C-Strings ❖ 245❖
cout<<“ The actual sizes of strings are: “;
cout<<sizeof(“Sun”)<< “ “<<sizeof (“OurEarth”)<<“ “<<sizeof(“Sky”) << “
“<<sizeof(“Mars”)<<endl;
return 0;}

The expected output is given below.

Size of Name = 36
The actual sizes of strings are: 4 9 4 5
The four names when written as a two dimensional array Name[4][9] occupy a total of 36
bytes.
In the second case, the names are stored by an array of pointers char* [4]. The array elements
are stored one after another separated by end of string null character (‘\0’). Thus the total size
is equal to 22 bytes. It is illustrated in Fig.10.2b below.

Name [0][9] S u n \0

Name [1][9] O u r E a r t h \0

Name [2][9] S k y \0

Name [3][9] M a r s \0

Fig. 10.2 (a): An array of strings. A two-dimensional array

char * name[4] = {"Sun", "OurEarth", "Sky", "Mars"}

S u n \0 O u r E a r t h \0 S k y \0 M a r s \0

Fig. 10.2 (b): An array of pointers to strings. There is no empty block

In programs which make a list of objects as elements of an array, remember that the first
entry is the 0th element in an array. If the output is also serially numbered starting from one,
do not forget to add 1 for printing of serial numbers.

10.6 STANDARD FUNCTIONS FOR HANDLING C-STRING CHARACTERS

C++ Standard library has a number of useful character handling functions which are listed and
explained in Table 10.1. The functions may be used for manipulations and tests on characters.
The applications of some of these functions are illustrated in Programs 10.17 to 10.20. All these
functions are of type int and receive a single integer argument. As per ASCII code characters are
also integers and in C++ they may be treated as constants with values as per ASCII code (see
Chapter 3) . Thus these functions manipulate characters as integers. If the functions are declared
❖ 246 ❖ Programming with C++

as char the return value is of type char. See the following program for islower() and
toupper()functions.

PROGRAM 10.17 – Illustrates application of islower() and toupper().

#include<iostream>
using namespace std;

void main()
{ char ch;
cout<<“Enter a sentence :\n” ;
while (cin.get(ch))

{ if ( islower(ch) )
cout<< char ( toupper (ch));
else
cout<<ch; }
}

The expected output is given below.

Enter a sentence :
Go to your class.
GO TO YOUR CLASS.

Table 10.1 – Standard functions for handling string characters

Function Description
int isalnum(int It returns true if c is a digit or alphabet, otherwise returns false.
c)
int isalpha(int It returns true if c is a letter, otherwise returns false.
c)
int isdigit(int It returns true if c is digit character, otherwise false.
c)
int islower(int It returns true if c is lower case letter, otherwise it returns false.
c)
int isupper(int It returns true if c is upper case letter, false otherwise.
c)
int tolower(int It returns lower case letter if c is upper case letter.
c)
int toupper(int It returns upper case letter if c is lower case letter.
c)
int isspace(int It returns true if c is any one of the white space character such as
c)
‘ ’, ‘\f ’, ‘\n’ , ‘\r’ , ‘\t’ , ‘\v’ ( see Chapter 2 for the descriptions) otherwise
returns false.
int isgraph(int c) It returns true if c is a printing character other than blank space ‘ ’,
otherwise returns false.
int isprint(int c) It returns true if c is a printing character or space ‘ ’ otherwise it returns
false.
int ispunct (int c) It returns true if c is a printing character other than a digit, a letter or
a space , otherwise it returns false.
int iscntrl(int c) If c is a control character such as ‘\n ’ , ‘\t ’ , ‘\r’ , ‘ \f ’, ‘\a’ , ‘\b’ the
function returns true otherwise false.
 C-Strings ❖ 247❖
Program 10.18 given below illustrates some of the above described functions.

PROGRAM 10.18 – Illustrates application of isalpha(), isdigit() and isalnum() functions.

#include<iostream>
using namespace std;

void main()
{ char ch;
int count =0;
int m =0, k = 0, n =0;

cout<<“Enter a sentence :\n” ;

while (cin.get(ch))
{ count ++;
if ( isalpha(ch) )
++m;
if( isdigit(ch))
k++;
if ( isalnum( ch))
n++;
if ( count ==25) break; }
cout<< “Number of letters = ” << m<<endl;
cout <<“Number of digits = ”<<k <<endl;
cout <<“Number of digits + letters = ”<<n <<endl;
}

The expected output is given below.

Enter a sentence :
Your number is 657876897.
Number of letters = 12
Number of digits = 9
Number of digits + letters = 21

10.7 STANDARD FUNCTIONS FOR MANIPULATION OF C-STRINGS

C++ Standard Library contains many useful functions for manipulation of strings. Table 10.2
gives the description of these functions. The manipulation of strings generally involves replacing
a string or substring with another string or substring, comparison of the two strings, appending
one string or substring to the front or back of another string, replacing n characters of one string
with n characters of another string etc. The applications of these functions are illustrated in the
programs given after the following table.
❖ 248 ❖ Programming with C++
Table 10. 2 – Standard C-string functions ( header file <cstring>)
For the sake of being brief we shall refer to strings pointed to by S1 and by S2 as strings S1 and S2.

Name of function and function arguments and description

char*strcpy (char *S1, const char *S2);
It copies S2 into S1 and returns S1. S1 becomes a copy of S2.
char*strncpy (char*S1, const *S2 , size_t n);
The function replaces the first n characters of S1 with first n characters of S2 and returns
S1. If n >= strlen(S2) then it behaves as strcpy(). size_t is the typedef of unsigned int.
char* strcat(char *S1, const char *S2);
The function appends the characters of S2 at the end of S1, returns S1. The terminating
null character of S1 is replaced by first character of S2.
char *strncat(char *S1, const char *S2, size_t n);
The function appends first n characters of S2 at the end of S1 and returns S1. The
terminating null character of S1 is replaced by first character of S2.
int strcmp(const char *S1,const char *S2);
Compares lexicographically string S1 with string S2 and returns – 1, 0, 1 if S1 is less
than , equal to or greater than S2 respectively.
int strncmp(const char*S1, const char *S2, size_t n);
Compares lexicographically the first n characters of S1 with first n characters of string S2
and returns – 1, 0, 1 if the substring of S1 is less than , equal to or greater than that of
S2 respectively.
size_t strlen (const char *S);
The function returns the length of string S ignoring the Null character at end of string.
char* strstr (const char *S1, const char * S2);
This function returns the address of first occurrence of S2 as substring of S1. It return 0
if S2 is not in S1.
char *strchr (const char *S , int C);
The function return pointer to first occurrence of C in S1, returns 0 if C not found.
char *strrchr (const char *S1, int C);
The function return pointer to the last occurrence of C in S1, returns 0 if not found.
size_t strspn (char *S1, const char* S2);
The function returns the length of substring of S1 that starts with the first character of
S1 and contains the characters of S2 only.
size_t strcspn ( char * S1, const char *S2);
The function returns the length of longest substring of S1 which begins with character
S1[0] and does not contain any character of S2.
char *strtok( char *S1, const char *S2);
S1 is tokenized into tokens delimited by characters of S2. It requires a number of calls
for complete tokenization. The first call is strtok(S1, S2); and in subsequent calls the
first argument is NULL. The subsequent call returns a pointer to the next token of S1.
 C-Strings ❖ 249❖

APPLICATIONS OF STRING MANIPULATING FUNCTIONS

strlen ( ) AND strcpy ()
The function strlen () returns the length of string (its argument), it includes white space but
ignores the end of string character (‘\0’).
The function call for strcpy is coded as below.
strcpy(S1, S2 );
where S1 and S2 are name of two strings. The function copies S2 into S1. In other words
S1 becomes a copy of S2. If S1 is longer than S2, it reduces it to the size of S2.

PROGRAM 10.19 – Illustrates strlen() and strcpy() functions.

#include<iostream>
# include <cstring>
using namespace std;
void main()
{
char Str1 [] = “Mona”;
char Str2 [] = “Learn C++”;
char Str3[] = “Lisa”;
cout<<“Lengths of Str1= ” <<strlen(Str1)<<endl;
cout<<“Length of Str2 = ” << strlen(Str2)<<endl;
cout<<“Contents of Str1 are = ”<<Str1<<endl;
cout<<“Contents of Str2 are = ”<<Str2<<endl;
strcpy(Str1, Str2);
cout << “Now contents of Str1 = ”<<Str1<<endl;
cout << “Now contents of Str2 = ”<<Str2<<endl;
}

The expected output is given below. The output is self explanatory.

Lengths of Str1= 4
Length of Str2 = 9
Contents of Str1 are = Mona
Contents of Str2 are = Learn C++
Now contents of Str1 = Learn C++
Now contents of Str2 = Learn C++

STRNCPY()
The function is called as illustrated below.
strncpy ( S1, S2, size_t n);
❖ 250 ❖ Programming with C++

Here S1, S2 are names of the two strings and n is a positive integer. The function replaces
the first n characters of S1 with first n characters of S2 and returns S1. If n is greater than the
length of S2 then the function behaves as strcpy(). The size_t is the typedef of unsigned int.
The following program illustrates its application.

PROGRAM 10.20 – Illustrates application of strncpy () function

#include <iostream>
# include <cstring>
using namespace std;
void main()
{ char Str1 [] = “John”;
char Str2 [] = “Mona Lisa”;
cout<<“Lengths of Str1 and Str2 are ”<<strlen(Str1)<<“ and ”<<
strlen(Str2)<<endl;
strncpy (Str1,Str2,4);

cout<<“After strncpy, contents of Str1 are = ”<<Str1<<endl;

cout<<“ After strncpy, Contents of Str2 are =”<<Str2<<endl;
cout <<“Length of str1 = “ <<strlen (Str1) << endl;
cout <<“Length of str2 = “ <<strlen (Str2) << endl;
}

The expected output given below shows that John in S1 has been replaced by Mona by the
operation of strncpy (Str1,Str2,4);.
Lengths of Str1 and Str2 are 4 and 9
After strncpy, contents of Str1 are = Mona
After strncpy, contents of Str2 are =Mona Lisa
Length of str1 = 4
Length of str2 = 9

FUNCTIONS STRCAT() AND STRNCAT()

The first function is called as below.

strcat(S1,S2 );
Here S1 and S2 are names of the two strings. The function appends the characters of S2 at
the end of S1. The application of this function is illustrated in the following program.
The second function is called as illustrated below.
strncat ( S1, S2, unsigned int n);
Here S1and S2 are the names of two strings and n is a positive number. The function appends
first n characters of S2 at the end of S1. However, if n is negative integer, some compilers may
show error while others may convert strncat() into strcat() and append the complete
S2 at the end of S1. If n is greater than the length of S2 then the function behaves as strcat().
The following program illustrates the two functions.
 C-Strings ❖ 251❖
PROGRAM 10.21 – Illustrates application of strcat()and strncat() functions.

#include<iostream>
# include < cstring >
using namespace std;
void main()
{
char Str1 [10] = “Mona” ;
char Str2 [14] = “Lisa Singh”;
char Str3 [10] = “Kirtika”;
char Str4 [10] = “Malhotra”;
strncat(Str1, Str2, 4);
cout<<Str1<<endl;
cout<<Str2<<endl;
strcat( Str3, Str4);
cout << Str3<<endl;
cout<<Str4 <<endl;
}

The expected output is given below.

Mona Lisa
Lisa Singh
Kirtika Malhotra
Malhotra

The first line of output is of Str1 after the operation of strncat(Str1,Str2, 4);.
The first 4 characters of S2 ( Lisa) are appended to end of Str1. The string Str2 is not affected
by this operation. The third line of output is the result of strcat(Str3,Str4);. The
complete string Str4 is appended to the end of string Str3. The string Str4 remains intact.

FUNCTION STRNCMP()

The function strcmp() is called as below.

strcmp(S1, S2);
It compares two strings S1 and S2 lexicographically and returns 1 if S1>S2 , returns 0 if S1
is equal to S2 and returns –1 if S1<S2. The following program illustrates the application of
the function.
PROGRAM 10.22 – Illustrates application of function strcmp()

#include<iostream>
# include < cstring >
using namespace std;
int main()
❖ 252 ❖ Programming with C++
{
char *Str1 = “XYZ” ;
char *Str2 = “ABC”;
char *Str3 = “XYZ”;
cout << strcmp(Str1, Str2)<<endl;
cout << strcmp(Str2, Str1)<<endl;
cout << strcmp(Str3, Str1)<<endl;
cout << strcmp(Str2, Str3)<<endl;
return 0 ;
}

The expected output is given below. Str1 is lexicographically greater than Str2 and Str1 is
lexicographically equal to that Str3. So the following output is obvious.
1
-1
0
-1

FUNCTION STRNCMP()

The function takes three arguments and is called as below.

strncmp(S1, S2, size_t n);
The function compares character by character the first n characters of string S1 with first n
characters of string S2. This is illustrated in the following program by making the two strings
different only by one character.
PROGRAM 10.23 – Illustrate application of function strncmp().
#include<iostream>
# include < cstring >
using namespace std;

int main()
{ char *Str1 = “You are making a drawing” ;
char *Str2 = “Xou are making a portrait”; // Y replaced by X
char *Str4 = “You are making a drawing” ;
char *Str3 = “Zou are making a picture”; // Y replaced by Z
cout << strncmp(Str1, Str2,15)<<endl;
cout << strncmp(Str2, Str1,15)<<endl;
cout << strncmp(Str3, Str1,15)<<endl;
cout << strncmp(Str2, Str3,15)<<endl;
cout << strncmp(Str4, Str3, 15)<<endl;
return 0 ;
}
 C-Strings ❖ 253❖
The expected output is given below. The output is obvious.
1
-1
1
-1
-1

FUNCTION STRSTR()

Function strstr() is called as strstr(S1, S2). The function returns the first occurrence of a substring
S2 in string S1. If S2 is not found it returns Null.

PROGRAM 10.24 – Illustrates application of strstr () function.

#include<iostream>
#include < cstring >
using namespace std;
void main()
{
char Str1 [ ] = “Lajpat Nagar” ;
char* ptr = strstr(Str1, “pat” ); //pat - sub string of Str1
cout<< ptr- Str1 <<endl; //pointer to pat minus pointer to
// to beginning of string. It will give index value.
cout << “Str1 [“<<ptr-Str1<<“]”<<endl;
cout<<Str1 <<endl;
}

The expected output is given below. See the above comment for explanation.
3
Str1 [3]
Lajpat Nagar

FUNCTIONS STRCHR() AND STRRCHR()

Functions are called as strchr(Str ,int C); and strrchr(Str , int C); Here Str is the name of string.
The function strchr () returns a pointer to the first occurrence of C in string Str while the
function strrchr() returns the pointer to the last occurrence of C. In the following program the
pointer to string which is the address of first element of string has been subtracted from the
pointer values given by the function so that it gives the index values where the C occurs.

PROGRAM 10.25 – Illustrates strchr() and strrchr()

#include<iostream>
# include < cstring >
using namespace std;
void main()
{
❖ 254 ❖ Programming with C++
char Str1 [ ] = “Ganga Nagar” ;
cout << “Index value because of function strchr(Str1,’g’ ):\n”;

char* ptr = strchr(Str1, ‘g’); //Application of strchr()

cout<< ptr - Str1 <<endl;

cout << “Str1 [”<<ptr-Str1<<“]”<<endl;
cout<<Str1 <<“\n\n”;

cout << “Index value due to function strrchr(Str1,’g’ ):\n”;

char* ptrr = strrchr(Str1, ‘g’ ); //Application of strrchr()
cout<< ptrr - Str1 <<endl;
cout << “Str1 [”<<ptrr-Str1<<“]”<<endl;
}

The expected output is given below. The first occurs at index value 3 while the last occurrence
happens at index value of 8.
Index value because of function strchr(Str1,’g’ ):
3
Str1 [3]
Ganga Nagar

Index value due to function strchr(Str1,’g’ ):

8
Str1 [8]

FUNCTION INT STRSPN( )

Function is called as int strspn(S1, S2). Here S1 and S2 are the names of two strings. The function
strspn() returns the length of the longest substring of S1 that starts with S1[0] and contains
only the characters found in S2.

FUNCTION STRTOK( )
An illustration of function call is given below.
char* T = strtok( Str2, “ ”);
According to this the function splits or breaks the string str2 into tokens. The splitting is
done at locations where a specified character or a substring is located or starts. In the above
declaration the tokenisation would occur at locations of white spaces. In other words the first
token is made on occurrence of first white space, i.e. after the first word. For continuing the
process we have to have NULL (‘\0’) as the first argument. This is illustrated in the following
program. For complete splitting the entire string into tokens we have to use a loop (while or
do—while, etc). In the following program a do – while loop is used.
 C-Strings ❖ 255❖
PROGRAM 10.26 – Illustrates application of int strspn() and strtok ()

#include<iostream>
# include < cstring >
using namespace std;
void main()
{ char Str1 [ ] = “I am making a picture ” ;
char Str2 [ ] = “He is making a portrait?”;
char S3[] = “I am making a drawing” ;
char S4[]= “You are doing portrait?” ;
cout << “Str1 = ” <<Str1 <<endl;
cout<<“Str2= ”<<Str2 <<endl;
int N13 = strspn(Str1, S3);
cout<<“N13 = ”<<N13<<endl;
int N14 = strspn(Str1, S4);
cout << “N14 = ” << N14<<endl;
char* K = strstr(S3,“a” );
cout << “K = ” << K<<endl;
char* T = strtok( Str2, “ ”);
int n =0;
do
{ ++n ;
cout << “\nToken ”<<n<<“ = ” << T ;
T = strtok( ‘\0’ ,“ ”); }
while ( T != ‘\0’); // do tokenize till the end of string
}

The expected output is given below.

Str1 = I am making a picture
Str2 = He is making a portrait?
N13 = 14
N14 = 0
K = am making a drawing

Token 1 = He
Token 2 = is
Token 3 = making
Token 4 = a
Token 5 = portrait?

10.8 MEMORY FUNCTIONS FOR C-STRINGS

These functions, in fact, deal with blocks of memory spaces which are treated as arrays of bytes.
❖ 256 ❖ Programming with C++

Therefore, any type of data stored on the bytes can be dealt with. The return values of functions
are void pointers. The arguments are also void pointers so that they can receive pointers to any
data type. In a way, these are templates. The different functions are described below. The return
pointer has to be cast into appropriate type before dereferencing because void pointers cannot
be dereferenced.
(i) void *memchr(const void * S, int C, size_t n);
Searches the first n characters of string pointed to by S for first occurrence of C. Returns
pointer to C in the string, if not found it returns 0.
(ii) void *memcpy (void * S1, const void* S2, size_t n);
Copies n characters of string pointed to by S2 on to string pointed to by S1 and returns
pointer to resulting string. The area from where the characters are copied and the area where
the characters are deposited are not allowed to overlap.
(iii) void *memcmp (const void* S1, const void *S2, size_t n);
Compares lexicographically the first n characters of S1 with those of S2. Returns 0 if
characters of S1 are equal, – 1 if less and +1 if greater than those of S2 .
(iv) void *memmove (void *S1, const void *S2, size_t n )
Copies n characters of S2 into S1 and returns pointer to the resulting string. The area from
where the characters are copied and the area where these are copied to are allowed to overlap.
(v) void *memset (void *S, int C, size_t n)
Copies ( unsigned char) C into first n characters of string pointed to by S, returns pointer
to resulting string.
The following two program illustrate the application of some of the above functions.

THE FUNCTION MEMCPY()

In the following program the function is called as below.

memcpy(Str1, Str2, 4);
In the operation of function, the first four characters of Str2, i.e. Mona are copied on to
first four characters of Str1 which is Lisa Lisa, so after the operation of the function Str1 becomes
Mona Lisa. The string Str2 is unchanged.

PROGRAM 10.27 – Illustrates application of memcpy() function.

#include<iostream>
# include < cstring >
using namespace std;
void main()
{ char Str1 [10] = “Lisa Lisa” ;
char Str2 [10] = “Mona” ;
memcpy(Str1, Str2 , 4);
cout<<Str1<<endl;
 C-Strings ❖ 257❖
cout<<Str2<<endl;
}

The expected output of the above program is given below.

Mona Lisa
Mona
The following program illustrates application of other memory functions.
PROGRAM 10.28 – Illustrates memchr (),memcmp(), memset(), memcpy () and
memmove ().

#include<iostream>
# include < cstring >
using namespace std;
void main()
{
char S1[32], S2 [] = “Learn C-strings.”;
char S3 [] = “Learn at proper time”;
char S4[] = “ABCDEFGH”;
char S5[] = “XYZ”;
/* Below static_cast<char*> is used to convert void pointer
into appropriate type.*/

cout << static_cast<char*>( memchr(S4, ‘D’,5))<<endl;

cout<<“memcmp( S2, S3 ,5) = ” <<memcmp( S2, S3 ,5) <<endl;

cout<< “memcmp ( S5, S4) = ” <<memcmp ( S5, S4, 5) <<endl;

cout<< “ String S4 after memset( S4, ‘R’, 4) is below.\n “;

cout<< static_cast<char*> ( memset( S4, ‘R’, 4))<<endl;
// copies the character ‘R’ into first 4 places of S4.
memcpy ( S1, S2, 33);
cout<<“\n”;
cout << S1 <<endl;
char S6[] = “Good morning John!”;
cout <<“S6 = ” << S6 <<endl;
cout<<“memmove( S6, &S6 [5], 12) is given below.”<<endl;
cout<<static_cast<char*> ( memmove( S6, &S6 [5], 12))<<endl;
cout <<“S3 = ” << S3<<endl;
cout<< “S1 = ” <<S1<<endl;
cout <<“memmove( S3, &S1[6], 20)) = ”;
cout << static_cast<char*> (memmove( S3, &S1[6], 20))<<endl;
}
❖ 258 ❖ Programming with C++

The expected output is given below.

DEFGH
memcmp( S2, S3 ,5) = 0
memcmp ( S5, S4) = 1
String S4 after memset( S4, ‘R’, 4) is below.
RRRREFGH
Learn C-strings.
S6 = Good morning John!
memmove( S6 , &S6 [5], 12) is given below.
morning John John!
S3 = Learn at proper time
S1 = Learn C-strings.
memmove( S3, &S1[6], 20)) = C-strings.

In the first line of output the substring after the letter ‘D’ is printed. For the second line of
output, the first 5 letters of S2, S3 are identical, so output is 0. For the third line, ‘X’ is
lexicographically greater than ‘A’ , therefore the output is 1. In the 4th line of output, 4 copies
of ‘R’ are copied on to S4 replacing ABCD. The result is obvious. In the 8-9 lines of output
the 12 letters of S6 starting from sixth element are copied and placed on first 12 characters. So
resulting string is morning John John!. In the last line of output the 20 characters of S1 starting
from seventh element are copied and placed on first 20 characters of S3, so resulting string is
C-strings.

EXERCISES
1. What are similarities and differences between arrays and C-strings?
2. What is the difference between “A” and ‘A’ in C++?
3. Declare a pointer to a string.
4. What is difference between cin.get() and cin.getline()?
5. How do you code the above two functions in a program to read up to 20 characters?
6. Make a program that reads user’s message like “Madan Mohan went to Mangalore on Monday.”
and counts the number of ‘e’, ‘M’ and number of total characters.
7. What is the difference between cout.put() and cin.putback()?
8. What does the function cin.peek() do?
9. Explain the differences between functions strchr() and strrchr().
10. What do you understand by the function strncat(S1, S2, unsigned int n)?
11. What do you understand by function strncpy(S1, S2,n)?
12. What does the following codes do with two strings with names S1 and S2 ?
(i) memcpy (S1, S2, size_t n);
(ii) strcpy (S1, S2);
(iii) strcat (S1, S2);
13. Give any two functions belonging to standard character handling library. How do you use them?
 C-Strings ❖ 259❖
14. How do you convert a string of digits into an integer value?
15. What the functions atol() and atof() do to a string of digits?
16. What is the difference between the functions strchr(Str, int C) and strrchr (Str,
int C)?
17. What do you understand by functions isupper( ) and tolower ( )?
18. Explain the working of function cin.ignore(). Make a program with cin.ignore (4) to
illustrate its working.
Answer:

PROGRAM 10.29 – An exercise to illustrate ignore().

#include<iostream>
using namespace std;

void main()
{char ch;
cout<<“Enter alphabets :\n” ;

while (cin.get(ch))
{cin.ignore(4);
cout<<ch;}
}

The expected output is as below.

Enter alphabets :
ABCDEFGHIJKLMNOPQRSTUVXYZ
AFKPU
19. Make a small program to illustrate the application of strlen() .
20. Make a program in which a user is asked to enter his/her name. The program should get it
verified by displaying the name and asking the user to enter yes if correct and no if incorrect.
21. Make a program which asks the user to enter one or more sentences and counts the numbers
of ‘e’ in the sentence.
Answer:

PROGRAM 10.30 – Counts the characters while reading.

#include <iostream>
using namespace std;
void main()
{
int count =0;
char para[150];
cout<<“Enter a sentence :\n” ;
cin.get(para,150) ;
❖ 260 ❖ Programming with C++
cout<<“you have written the following.\n”<<para<<endl;
for (int i = 0; i<150;i++)
if ( para[i] ==’e’ ) ++count;
cout<<“There are ”<<count<<“‘e’ letters in it.”<<endl;
}

The expected output is as under.

Enter a sentence :
Here every thing is at higher price.
you have written the following.
Here every thing is at higher price.
There are 6 ‘e’ letters in it.
22. Make a Program asking the user to write a sentence with digits. The program counts the
number of digits in the sentence.
Answer:

PROGRAM 10.31 – Illustrates counting only digits in characters read.

#include <iostream>
using namespace std;
void main()

{
int count =0;
char para[150];
cout<<“Enter a sentence with digits: \n” ;
cin.get(para,150) ;

cout<<“\nyou have written the following sentence.\n” <<para <<endl;

for (int i = 0; i<150;i++)
if ( para[i] >=’0' && para[i]<=’9') ++count;

cout<< “\n There are ” <<count <<“ digits”<<“ in this.”<<endl;

}

The expected output is as under.

Enter a sentence with digits:
Add number 564 to 342 and subtract 64 from sum.

you have written the following sentence.

Add number 564 to 342 and subtract 64 from sum.

There are 8 digits in this.

❍❍❍
CHAPTER

11.1 INTRODUCTION

A major feature which distinguishes C++ from C is the concept of classes and object oriented
approach to programming. A class is a method of organizing the data and functions in the same
structure which describes the properties or characteristics of real life objects. It is an extension
of structure of C, but C-structures (discussed at the end of this chapter) have many limitations.
One limitation is that it has only public members. The classes of C++ have the provision of
categorizing the members into private, protected and public and thus they allow information
hiding. Besides, the classes support inheritance, i.e. new classes may be derived from an existing
class. The derived class uses the existing class as it is, besides, it may add new data and functions
which add new features to the objects of the derived class. The process of inheritance is endless.
You may derive newer classes from derived classes. Thus a very big program may be thought of
consisting of several classes with objects that serve the purpose of the program. Each class may
be developed independently, debugged and tested before putting in the big program. In this
chapter, we shall introduce the concept of classes, constructor, destructor functions, local classes,
structures etc., while the advanced topics are discussed in Chapter 12.
We can think of a class as a blueprint or a design in programming with abstract data
variables and the relevant functions that describe the common characteristics of a group of
objects. The functions of class operate on object data to yield useful information about the
object of the class. The class by itself does not do anything. The class functions cannot be
called on their own. They are called by linking them with the objects of class on whose data
they operate. For instance, you cannot ride a car which is just a design on the paper. You can
ride only when a car is built. Similarly a class program works only when we create an object
or objects of the class. Also, once the car design is done, same design can be used to produce
a large number of cars. In the same way, once a class code is perfected one can create as many
objects of the class as desired and they can all co-exist in the memory. A class contains data
variables and functions which are called data members and function members respectively.
The various class functions operate only when linked with an object of that class. That is
why the name object oriented programming (OOP). There are, however, different opinions
on what comprises the object oriented programming. The other attributes of classes are also
part of OOP. The majority of programmers agree that OOP includes data abstraction,
encapsulation of data and function members, information hiding, inheritance, polymorphism
❖ 262 ❖ Programming with C++

and dynamic binding. While some are of the opinion that OOP also includes generic
programming. All these terms would be explained in this and subsequent chapters. The
function members of a class generate information relevant to the particular object with whose
identifier the functions are called. For example, in a computerised company the employees are
allowed to check the entries in their pay accounts on their terminals. Let display_pay()
be a function of the class program which the employee can invoke to get to their pay accounts.
The employee is an object of the class and has an identifier in form of employee code, we
call it employee_identifier. The employee would enter employee code and would ask
for display as illustrated below.
employee_identifier.display_pay();
Notice that dot operator (.) provides link between the employee code and the function
display(). If the employee simply calls the function without the employee identifier and
the dot operator, i.e. display_pay(); it will not operate, i.e. it will not display the pay
entries of any employee. In a class, the operation of a function member is tied to the objects
of the class.
The characteristics of objects of the physical world may be described by class programs.
We all know cats belong to a different class of animals from that of dogs but both belong
to the class animal. Let us name these classes as class Cat, class Dog and class Animal. Evidently,
class Cat and class Dog may be inherited from class Animal, because, both the cats and dogs
have some characteristics which are common with other animals as well. Also any one dog say
mydog is one object or instance of the class Dog. In the same way if we think of manufactured
goods, we can easily make a class for cars another for scooter, still another for vehicle. We may
prepare data and functions to describe their characteristics. That is to say there are some
common characteristics among all the scooters, similarly cars have common characteristics
among themselves. Also some characteristics may be common with all other vehicles. For
instance all vehicles have an engine, a registration number, etc. So it is possible to have a base
class which we may call class Vehicles and the class Cars and the class Scooters as derived classes
as illustrated in the figure below. That explains inheritance.

class Vehicles class Animals

class Cars class Scooters class Dogs class Cats

Fig. 11.1: Classes and inheritance

11.2 DECLARATION OF CLASS AND CLASS OBJECTS

The word class is a keyword of C++ and for declaring a class it is essential to write class followed
by identifier or name of the class. In the declaration class Car for cars, class Scooters for scooters
and class Vehicles for vehicles the word class is the keyword and Cars, Scooters and Vehicles are names
of the respective classes. A short outline of a class declaration is given below.
 Classes and Objects–1 ❖ 263❖
class identifier
{access specifier: // Body of class having data members
statements; //and functions members enclosed by curly braces {}.
}; //The class declaration ends with the semicolon.
The class body comprises data members and function members. The data members
comprise abstract data that applies to all the object of class and is relevant to the aim of the
program. The common characteristics of objects of class are described by the function members
of the class.
For the class Car, a particular car say mycar becomes an instance or an object of the class
Cars if it is so declared. It (mycar) will have particular set of data. Similarly other cars may
be declared as objects of the class Car and will have their own but similar sets of data. The
objects of a class carry individual copies of their data unless data is declared static in which
case all objects share same copy of data. The data of mycar becomes the arguments for the
function members of the program class Cars when they are called for the object mycar. But
before an object say mycar can call a function of class Cars we have to declare that mycar is
an object of the class - class Cars. You remember when we declare a variable we first mention
its type such as int, float, double or character. Similarly the type of an object of a class is the
name of its class. For instance, mycar is an object of class Cars, therefore, its type is Cars. With
declaration of class no memory is allocated. Memory is allocated when class objects are defined.
Now an object belonging to the class Cars may be defined as below.
class Cars mycar; //declaration of object
Or we may simply write it as below.
Cars mycar; //Cars is the type for the objects of Cars.

Name of class name of object

If more than one objects belonging to same class say class Cars such as car1, car2, car3 are
to be declared it may be done as below,
Cars car1, car2, car3 ;

Name of class Names of objects separated by comma

With the above definition of the objects, each object will be allocated memory of same
size which is appropriate according to declarations in the class. Objects may also be declared
by simply adding names of objects after the closing right brace of class body and before the
semicolon.
The class contains data members and function members which apply to all the objects of
the class. The choice of the functions and data is done according to the purpose of the class
program. For example, a class program called class CarM for a car manufacturer may have
different data and functions from that of the class CarD for the dealer of car, though both the
classes may deal with same cars. The detailed declaration of class is illustrated below.
❖ 264 ❖ Programming with C++

DECLARATION OF CLASS
The form of code used for declaring a class is illustrated below.
class X // class is keyword, X is the name of class
{
private: //access label private
memberA; // memberA is a private member of class X.
protected: // access label protected
memberB; // MemberB is protected member
...... // dotted line stands for other data or functions
public: //access label public
......
memberC; //memberC is a public member
———
private: //a access label may be repeated
memberD; // memberD is also a private member of class X
}; // end of class

Starting from the top of the above declaration, class is the keyword of C++ and is must
for every class declaration. The identifier X is the name of the class. Name may be chosen as
we do for any other variable (see Chapter 3). The body of class begins with left brace ‘{’ after
the class name and ends with the right brace ‘}’ followed by semicolon (;) at the end. Three
access labels are permitted. These are private, protected and public. These access labels along
with their respective functions may be written in any order in the class body. Also any access
specifier may appear more than once in the class body. You may write first the access label
private: followed by private data and functions members, the second may be public: followed
by public data and function members. Note that every access label ends with a colon (:).
If there is no access specifier at the start, it is taken that the data members and functions
up to next access label are private by default. The function members of a class may be defined
inside or outside the class body. If functions are defined outside the class body their prototypes
must be declared in the body of class.
The initialization of data declared private is carried out through public functions which may
also be the constructor functions of the class. The constructors are discussed in Section 11.8.

11.3 ACCESS SPECIFIERS – private, protected and public

PRIVATE:

The class members written under the access label private are accessible by class function members
or friend functions and the functions of friend classes. The keyword friend provides this special
privilege to the function or class declared as friend in the class body. In Chapter 12 we shall deal
with friend functions and friend classes. The data or functions declared private cannot be accessed
 Classes and Objects–1 ❖ 265❖
from outside the class. The objects of the class can access the private members only through the
public function members of the class. This property of class is also called information hiding.
However, with the help of pointers one can indirectly access private data members provided the
class contains at least one public data member. Generally data members are declared private. The
intention is to hide as much information about data and implementation of class as possible.
This also helps against accidental change in object data members. The functions members are
in general declared public.

PROTECTED:

The class members declared protected are accessible by the function members of same class and
friend functions and friend classes as well as by functions of derived classes. For all other
operations protected members are similar to private members.

PUBLIC:

The class members declared public are accessible from inside or outside the class. Some of the
public functions of class provide interface for accessing the private and protected members of the
class. For instance, initialization of a class object with private or protected data is always carried
through public functions. Some public function also provide interface for other classes.
Figure 11.2 below gives a physical picture of private, protected and public members of a class.
Program 11.1 given below, illustrates a simple class with only one public function.

Karan
Karan

Fig.11.2: A physical illustration of public and private members of a class

PROGRAM 11.1 – Illustrates a class declaration and function definition in the class.

#include <iostream>
using namespace std;

class Market // Market is name of class.

{ //The ‘{’ marks beginning of class declaration.
public: // Access label public
void Display() // Definition of a public function Display ()
{ cout<< “Welcome To This Market” << endl;
cout << “Here is the price List” <<endl;}

}; //Semicolon after ‘}’ marks end the of class declaration.

❖ 266 ❖ Programming with C++
void main() // start of main program.
{Market L1; // declaration of object L1 of class Market
L1.Display(); // Function linked with object L1
} // dot(.) is selection operator

The expected output is as under.

Welcome To This Market
Here is the price List
Note that class declaration is separate from the main program which, in the above case,
starts with void main(). There is only one function void display () defined in the class body.
In the main program we first declare an object of class by name L1, i.e. Market L1; . The
function display() is called by writing L1.Display (); The dot operator (.) links
the function display with object data of L1. In this example there is no data but below you
will have more examples.

11.4 DEFINING A MEMBER FUNCTION OUTSIDE THE CLASS

The above program is repeated below with the prototype of member function declared in the
class but function is defined outside the class body. The output is of course the same as given
above. For defining a function outside the class body one has to use the scope resolution operator
(::). An illustration is given below.
type class_name :: Function_name (type parameter1, - - -, - - -, type parameter n)
{ statements ; } // function body
Here type refers to the type of the data returned by the function, it is followed by name
of the class, then the scope resolution operator followed by function name and parameters in
brackets. This is followed by body of function in braces{}.

PROGRAM 11.2 – Illustrates function definition outside the class.

#include <iostream>
using namespace std;
class Market
{ // start of class declaration
public:
void Display(); // function prototype
}; // end of class
/*In the following function definition, void is type of function, Market
is class name, :: is the scope resolution and Display is name of function, there
are no parameters*/
void Market::Display () //function definition outside the class
{cout<< “Welcome To This Market” << endl;
cout << “Here is the price List” <<endl;}
 Classes and Objects–1 ❖ 267❖

void main() //main program

{Market L1; // L1 is declared an object of class market
L1.Display(); // L1 calls function display
}

11.5 INITIALIZING PRIVATE DATA MEMBERS

An object of a class is constructed when its data members are initialized. If data is declared private,
the initialization is carried by a public function which may be a constructor function (described
in Section 11.8). In Program 11.3 given below, a class has both the public members and private
members. The data members x, y and z are declared private. For accessing the private data
members we need a public function. This is provided as Setprice (). In the class body, it is defined
as below.
void Setprice (int a, int b, int c)
{x = a,y = b, z = c;}
For defining the same function outside the class body, first we have to declare its prototype
in the class as,
void Setprice (int , int , int );
and put the definition outside the class as given below.
void Market ::Setprice (int a , int b , int c )
{x = a, y = b , z = c ;}
The general practice is to declare the function prototype inside the class and to define the
same outside the class. The difference between the two is that functions defined inside the
body of class are taken as inline functions while the ones defined outside the class body are
called at the time of execution. The functions defined outside may also be declared inline by
keyword inline.

PROGRAM 11.3 – Illustrates initialization of private data with a public function.

#include <iostream>
using namespace std;
class Market
{
private :
int x,y,z;
public:
void Display1()
{cout<< “Welcome! Here is the price List ” << endl;}
void Setprice (int a, int b, int c) // function defined in class
{x = a, y = b, z = c;}
void Display2 ()
{cout<<“ Price of item1 = ”<< x<<endl;
❖ 268 ❖ Programming with C++
cout<<“ Price of item2 = ”<< y<<endl;
cout<<“ Price of item3 = ”<< z<<endl;
}
}; // End of class
void main()
{
Market L1;
L1.Setprice (4,6,2); // Object L1 is initialized by function
// setprice()
L1. Display1(); // Functions called by using dot operator.
L1.Display2();
}

The expected output is given below. The output is self explanatory.

Welcome! Here is the price List
Price of item1 = 4
Price of item2 = 6
Price of item3 = 2
In the Programs 11.4 and 11.5 we try to bring out the significance of declaring the data
private. In Program 11.4 we declare a class Rect which deals with rectangles. It has one public
function by name Area which finds area of rectangle if sides are given. The data variables are
declared public. The function area is also public. R1 is an object of class Rect. You would see
that R1 can directly access the data variables and can assign values as done in the main() by
the following statements.
R1.x = 8;
R1.y = 5;
The program illustrates that the public data members are directly accessible by objects of
class.

PROGRAM 11.4 – Illustrates initializing public data members from outside the class.

#include <iostream>
using namespace std;
class Rect { // beginning of class
public:
int x ,y; // data declared public
int Area( )
{return x*y; }
}; // End of class
int main()
{
Rect R1; //R1 is an object of class Rect
 Classes and Objects–1 ❖ 269❖
R1.x = 8; // accessing data members directly
R1.y = 5;
cout << “Area = ”<< R1.Area()<<endl;
cout << “Length = ” << R1.x <<endl;
cout<< “Width = ” << R1.y << endl;
return 0 ;
}

The expected output given below shows that object can directly access the data declared public.
Area = 40
Length = 8
Width = 5

ACCESSING PRIVATE DATA MEMBERS

The above program is repeated below, now with the data x, y declared private. Now x and y are
not accessible by the object directly and we have to create a public function to initialize the object
data so that rectangle is realised. This is done by public function void setsides (). In
this function we declare two int variables L and W and we equate x = L and y = W. Since L and
W are declared in the public domain they are accessible to the objects of class as already shown
in Program 11.4 above. So in the main () we create an object R1 of the class and in next line
the function setsides()is called and the values 8 and 5 are put in as its arguments. These are
the values of L and W which are passed on to x and y through the function setsides(). These
values of x and y are used to calculate the area. The Program runs to give the area 40.
Now if you try to access the data of the rectangle by calling R1.x or R1.y as done in the
previous Program 11.4 you would get the following error message.
‘x’ : cannot access private member declared in class ‘Rect’
‘y’ : cannot access private member declared in class ‘Rect’
This is so because data is private. If it is required to display data you have to have another
function in public domain to display the length and width of rectangle. This is carried out
in the following program for sake of illustration. The following program shows that private
data can be extracted through public functions.

PROGRAM 11.5 – Illustrates accessing private data with public function.

#include <iostream>
using namespace std;
class Rect {
private :
int x ,y;
public:
void setsides ( int L, int W){ x = L, y = W;}
❖ 270 ❖ Programming with C++
int Area( ){return x*y;} // definition of function Area ()
// public function to get object data
void Dimension (){cout <<“Length = ”<< x <<“,\tWidth = ”<<y << endl; }
}; // End of class
int main()
{
Rect R1;
R1.setsides (8,5) ; //Calling setside () to initialize x,y
cout << “Area = ”<< R1.Area()<<endl;
R1.Dimension (); // calling function Dimension ()
return 0 ;
}

The expected output is given below.

Area = 40
Length = 8, Width = 5
An important feature of class programming is its ability to hide private data and private
functions. Therefore, in general, the intention is to hide as much as is feasible without
impairing the normal execution of the program.

11.6 CLASS WITH AN ARRAY AS DATA MEMBER

The class data or functions may have arrays. In the following program the private data is an array.
The functions void Display1() and void Display2()are defined inside the class while
the function Setprice() is defined outside the class. As required, its prototype (given below)
is declared inside the class.
unsigned int n;
void Setprice (int x[], n);

PROGRAM 11.6 – Illustrates how to access private array data member.

#include <iostream>
using namespace std;
class List {
private : //Access label private
int x[4]; // Array data
public:
void Display1() // Definition of function Display1()
{cout<< “Welcome To This Market” << endl;
cout << “Here is the price List” <<endl;}
void Setprice (int x [4] );
void Display2() // Definition of function Display2()
 Classes and Objects–1 ❖ 271❖
{for (int i = 0; i<4; i++)
cout<<“ Price of item ” ”<<i+1<< ” = “<< x[i]<<endl; }
}; // End of class
void List ::Setprice (int a[4] ) // Definition of Setprice ()
{ for ( int j = 0; j< 4; j++)
x[j] = a [j] ; }
void main()
{
List L1; // L1 is an object of class List
int P[4] = {6,5,4,8} ;
L1.Setprice(P) ; // P is const pointer to the array
L1. Display1();
L1.Display2();
}

The expected output is given below.

Welcome To This Market
Here is the price List
Price of item 1 = 6
Price of item 2 = 5
Price of item 3 = 4
Price of item 4 = 8

11.7 CLASS WITH AN ARRAY OF STRINGS AS DATA MEMBER

In the following program the private data includes an int array which represents marks obtained
by students and an array of C-strings for entering the names of students. The aim is to prepare
a list in which marks are written against individual student name.

PROGRAM 11.7 – Illustrates how to access private array and string data.

#include <iostream>
#include<string>
using namespace std;
class List
{
private :
int x[4]; // Array data
char Names[4][20] ; // two dimensional array
public:
void Display1()
{cout << “Here is the Marks List” <<endl;}
❖ 272 ❖ Programming with C++

void Setmarks(int x [4] ); // Declaration of function

void Setnames (char Names[4][20] ); // Declaration of function
void Display2() // Function defined inside the class.
{for (int i = 0; i<4; i++)
cout<<“ Marks obtained by ”<<Names [i]<<“ = ”<< x[i]<<endl;}
}; // End of class
void List :: Setmarks (int a[4] ) //Definition of setmarks ()
{for ( int j = 0; j< 4; j++) // outside the class
x[j] = a [j] ;}
void List:: Setnames(char b[4][20] ) //Definition of Setnames
{for ( int m = 0 ; m<4; m++) // outside the class
{for (int p = 0 ; p<20; p++)
Names[m][p] = b[m][p] ;}
}
void main()
{
List L1;
char Students [4] [20] ;
cout<<“Enter the names of 4 students: ”;
for( int i =0; i< 4;i++)
cin.getline ( Students[i],20); // Reading names from keyboard

cout<<“Enter Marks.”<< endl;

int P [4];
for (int k =0; k<4;k++)
cin>> P[k];
//L1 is an object of class
{L1.Setmarks(P) ; // Putting in an array parameter
L1.Setnames (Students);} // Putting in string parameter
L1.Display1();
L1.Display2();
}

The program output is given below.

Enter the names of 4 students: Anjali
Sujata
Sunita
Mohan
Enter Marks.
76 68 87 75
 Classes and Objects–1 ❖ 273❖
Here is the Marks List
Marks obtained by Anjali = 76
Marks obtained by Sujata = 68
Marks obtained by Sunita = 87
Marks obtained by Mohan = 75

The scheme of the program 11.7 is extended in the following program. Here marks
obtained in 4 subjects are listed against student names. The data members for the marks and
names of students are declared private and are accessed through the member functions
Setmarks() and Setnames () respectively. The program handles an array of strings in Setname()
and four int arrays in Setmarks().
In this program the array MA [4] represents marks in mathematics, PH[4] represents
marks in physics, CH[4] stands for marks in chemistry, ENL[4] for marks in English and Total
represents the total of marks in mathematics, physics, chemistry and English for each student.
The student names are represented by an array of strings Names[4][20] which is used to
represent 4 names each having a maximum of 20 characters, this includes the null character.
The output gives the students name and marks obtained in different subjects and total in a
single row.

PROGRAM 11.8 – Illustrates the preparation of mark list in 4 subjects and total marks.

#include <iostream>
#include<string>
using namespace std;
class List
{ int MA[4], PH[4], CH[4],ENL[4], Total[4]; // private by default
char Names[4][20] ;
public:
void Display1()
{cout << “Here is the Marks List” <<endl;}
void Setmarks(int MA [4], int PH[4], int CH[4], int ENL[4], int Total[4]);
void Setnames (char Names[4][20]);
void Display2()
{cout<< “Students/Subjects Math PHYS CHEM ENGLISH Total
“<<endl;
for (int i = 0; i<4; i++)
{Total[i] = MA[i] + PH[i] + CH[i] + ENL[i];
cout <<Names [i]<<“ \t \t” << MA[i]<<“ \t “<<PH[i]<<“ \t “<<
CH[i]<<“ \t “<<ENL[i]<<“ \t “ << Total[i]<<endl;
} } }; // end of class
void List::Setmarks(int a[4],int b[4],int c[4],int d[4],int e[4])
{for (int j = 0; j< 4; j++)
{ MA[j] = a [j];
PH[j] = b[j];
❖ 274 ❖ Programming with C++
CH[j] = c[j];
ENL[j] = d[j];
Total[j] = e[j];} }

void List:: Setnames(char b[4][20] ) // definition of Setnames

{ for ( int m = 0 ; m<4; m++)
{for (int p = 0 ; p<20; p++)
Names[m][p] = b[m][p] ;}
}

void main() // start of main program

{
List L1;
char Students [4] [20] ;

cout<<“Enter the names of 4 students: ”;

for( int i =0; i< 4;i++)

cin.getline ( Students[i],20);

int M[4];
cout <<“Enter Marks in Math”<<endl;
for ( int k =0; k<4 ; k++)
cin>>M[k];

int P[4] ;
cout <<“Enter Marks in Physics”<<endl;
for ( int m =0; m<4 ; m++)
cin>>P[m];

int C[4] ;
cout <<“Enter Marks in Chemistry”<<endl;
for ( int s =0; s<4 ; s++)
cin>>C[s];

int E[4] ;
cout <<“Enter Marks in English”<<endl;

for ( int n =0; n<4 ; n++)

cin>>E[n];
int T[4];
{L1.Setmarks(M, P, C, E, T ) ;
L1.Setnames (Students);}

L1. Display1();
L1.Display2();
}
 Classes and Objects–1 ❖ 275❖
The expected output is given below.
Enter the names of 4 students: Sujata
Sunita
Namrita
Aakriti
Enter Marks in Math
67 76 65 87
Enter Marks in Physics
67 87 66 96
Enter Marks in Chemistry
88 76 85 76
Enter Marks in English
65 67 68 75
Here is the Marks List
Students/Subjects Math PHYS CHEM ENGLISH Total
Sujata 67 67 88 65 287
Sunita 76 87 76 67 306
Namrita 65 66 85 68 284
Aakriti 87 96 76 75 334

The above program may be extended to any number of students and any number of
subjects or tests by making corresponding changes in the number of elements in the arrays.
In the following example, we declare a class with name Cubicle. The name stands for any
rectangular prismatic solid body having three dimensions, i.e. length, width, and height. The
program calculates the surface area and volume of cubic prisms. Two instances or objects are
created by names cube1 and cube2. Since their type is Cubicle, so in the declaration they are
written as Cubicle cube1; and Cubicle cube2; The functions are called by the objects
with the help by selection dot operator (.). The link between the object and the function is
through the pointer this which is generated by compiler for every object that is created for a
class. The dot operator passes on the pointer to the function and thus the function operates
on object data. (for more details on pointer this see Chapter 12)

PROGRAM 11.9 – Illustrates another application of class program.

#include <iostream>
using namespace std;
class Cubicle { // Cubicle is name of class
private:
int x,y,z;

public:
void set_sides( int , int, int );
int volume (); // declaration of function volume ()
int surface_area (); // declaration of function surface_area ()
}; // End of class
❖ 276 ❖ Programming with C++
void Cubicle:: set_sides (int width, int height, int length )
{ x = width , y = height, z = length ; }
int Cubicle::surface_area() //Function surface_area defined
{ return 2*(x*y+x*z+y*z);} //outside the class

int Cubicle::volume () //Function volume defined outside

{ return x*y*z;} // the class

int main()
{ Cubicle cube1; //cube1 is an object of the class Cubicle.
Cubicle cube2 ; // cube2 is another object of the same class.
cube1.set_sides (3,4,5); // The dot (.) is a selection symbol.
cube2.set_sides (2,4,6); // set_sides initializes the object

cout << “Volume of cube1 = ” <<cube1.volume()<<“\n”;

cout<<“Surface area of cube1 = ”<<cube1.surface_area()<<“\n”;

cout<< “Volume of cube2 = ” <<cube2.volume()<<“\n”;
cout<<“Surface area of cube2 = ”<<cube2.surface_area()<<“\n”;
return 0 ;
}

The expected output is as below.

Volume of cube1 = 60
Surface area of cube1 = 94
Volume of cube2 = 48
Surface area of cube2 = 88

In the above example, class Cubicle, the dimensions x, y and z are declared private
while the functions set_sides (), volume() and surface_area() are declared public
members of the class. Also notice that these functions are defined outside the class, therefore,
they are prefixed with name of class and scope resolution operator (::). For example, for
defining volume function, the first line of code is int Cubicle ::volume(). In the
main() two instances are declared as Cubicle cube1; and Cubicle cube2 ; Both
are preceded by Cubicle because it is their type. Both the objects may be declared in single
line, in that case it is coded as Cubical cube1, cube2;It is just like we declare two
integers as int x, y;. The data members are accessed through the public function set_sides
(). For cube1 it is coded as cube1.set_values(3,4,5);. A similar statement is written
for cube2 which is another object of the class. In the output statements also we have to use
selection symbol (.) for calling the class functions, for instance, cube1.volume() and
cube2.volume() etc. to indicate that here we want to have the volume of cube1 and here
we want to have volume of cube2, etc. The output is easily verified by manual calculations.
 Classes and Objects–1 ❖ 277❖

11.8 CLASS CONSTRUCTOR AND DESTRUCTOR FUNCTIONS

In the previous programs the private data members of the classes have been accessed through a
public member functions such as void set_sides () of Program 11.9. These functions are called
after the objects have been created. It is not instantaneous initialization of the object at the time
of declaration of object in the way we declare and initialize a variable like
int x = 6;. In case of objects also the same may be achieved with a constructor function
which is a special public function of the class.
A constructor function is a public function and has same name as that of the class.
It is used to initialize the object data variables or assign dynamic memory in the process of
creation of an object of the class so that the object becomes operational. Whenever a new
instance or an object of the class is declared the constructor is automatically invoked and
allocates appropriate size of memory block for the object. The constructor function simply
initializes the object. It does not return any value. It is not even void type. So nothing is written
for its type. The general characteristics of constructor functions are given below.
(i) Constructor function has same name as that of the class.
(ii) It is declared as a public function.
(iii) It may be defined inside the class or outside the class. If it is defined outside the class
its prototype must be declared in the class body.
(iv) It does not return any value, nor it is of type void. So no type is written for it.
(v) The constructor is automatically called whenever an object is created.
The destructor function removes the object from the computer memory after its relevance
is over. For a local object the destructor is called at the end of the block enclosed by the pair
of braces { } wherein the object is created and for a static object it is called at the end of main()
function. It releases the memory occupied by the object and returns it to heap. The destructor
function also has same name as class but it is preceded by the tilde symbol (~) and has no
parameters. It is a good programming practice to provide a destructor explicitly. If it is not
provided explicitly in the program it is provided by the system.
The following program illustrates the parametric constructor. In such cases, objects are
declared along with the object data values which are passed on as arguments of functions.

PROGRAM 11.10 – Illustrates a class constructor.

#include <iostream>
using namespace std;
class Cubicle {
private:
int x, y, z;
public:
Cubicle ( int , int, int ); //constructor prototype
❖ 278 ❖ Programming with C++
int volume() {return ( x*y*z);}
}; // End of class
Cubicle::Cubicle(int a, int b, int c ) // Constructor
{x = a, y = b, z = c ; // defined outside class
cout<<“constructed called”<<endl;}

int main()
{Cubicle cube1(3,3,3) ; // object 1 with arguments 3,3,3
Cubicle cube2(4,5,6); //object 2
Cubicle cube3 (2,4,5); //object 3

cout << “ Volume of cube1 = ” <<cube1.volume()<<“\n”;

cout << “ Volume of cube2 = ” <<cube2.volume()<<“\n”;
cout << “ Volume of cube3 = ” <<cube3.volume()<<“\n”;
return 0 ;
}

The expected output is given below.

constructed called
constructed called
constructed called
Volume of cube1 = 27
Volume of cube2 = 120
Volume of cube3 = 40

From the output it is clear that every time an object is created the constructor function
is invoked. In the above program three objects cube1 and cube2 and cube3 are created. For
the three objects the constructor has been called three times. The destructor function is called
when the object goes out of scope.

11.9 TYPES OF CONSTRUCTORS

A program may have more than one constructor functions provided their arguments are different.
At least one parameter should be different or at least of different type. The different types of
constructors are listed below.
(i) Constructor with default values
(ii) Constructor with parameters
(iii) Copy constructor
The parametric type of constructor has already been illustrated in the Program 11.10
described above. The following program illustrates default constructor, parametric constructor,
copy constructor and destructor functions.
In the following program we illustrate the different types of constructors and how these
are invoked when objects with different type of arguments are called.
 Classes and Objects–1 ❖ 279❖
PROGRAM 11.11 – Illustrates types of constructors and destructor.

#include <iostream>
using namespace std;
class Cubicle {
private:
int x, y, z;
public:
Cubicle () { x = 3, y = 4, z = 2 ; // default constructor
cout << “Default constructor called.”;}
~ Cubicle () {cout<< “Destructor called to remove object.\n”;}
//Following is the parametric constructor definition
Cubicle (int a, int b, int c ){x = a, y = b , z= c;
cout << “Parametric constructor called. ”;}
Cubicle (Cubicle & cubeA) // copy constructor definition
{ x = cubeA.x , y = cubeA.y , z = cubeA.z ;
cout << “\nCopy constructor called.” ;}
int volume() {return ( x*y*z);} // definition of function volume
}; // End of class
int main()
{{ Cubicle cube1; //scope of cube1 is this program block{}
cout << “\nVolume of cube1 = ” <<cube1.volume() <<“\n” ;}
Cubicle cube2(6,5,4) ; // Scope main()
cout << “\nVolume of cube2 = ” <<cube2.volume();

{ Cubicle cube3 (cube2) ; //Scope of cube3 is this block {}

cout << “\nVolume of cube3 = ” << cube3.volume( )<<“\n”;}
return 0 ;
}

The expected output is given below.

Default constructor called.
Volume of cube1 = 24
Destructor used to remove object.
Parametric constructor called.
Volume of cube2 = 120
Copy constructor called.
Volume of cube3 = 120
Destructor called to remove object.
Destructor called to remove object.

In the above program object cube1 is declared without any argument so the compiler
chooses the default constructor for its construction. The object is created in a block contained
❖ 280 ❖ Programming with C++

within curly brackets. The scope of this object is that block of program in which it was
created. At the end of block destructor was called to remove it from the memory. See third
line of output.
Object cube2 is created in main () so its scope is up to end of main(). Object cube3 is
a copy of cube2, so copy constructor is called to initialize it. The scope of object cube3 is also
up to the end of block in which it is created. So at end of block the destructor is called to
remove it from the memory. At the end of main the destructor is again called to remove the
object cube2.
In case of default constructors some values may be by default while others may be
parametric. So there could be variety of constructor types. Thus there could be constructors
with all default argument, constructors with one parametric argument and other default,
constructors with two parametric arguments and other default and so on. The following
program illustrates the application of some of these constructor functions.

PROGRAM 11.12 – Illustrates types of constructors.

#include <iostream>
using namespace std;
class Cubicle {
private:
int x, y, z;
public:
Cubicle () { x = 3, y = 4, z = 2 ;}
//above is constructor with default values
~ Cubicle () {cout<< “Destructor called to remove object.\n”;}
Cubicle (int a){x = a, y = 2 , z= 3; //one parameter constructor
cout << “\nConstructor with one parameter used. ”;}
// below is two parameter constructor
Cubicle (int m,int k){x = 3,y = m, z= k;
cout<<“\nconstructor with two parameters used.”; }
Cubicle (Cubicle & cube2) // copy constructor
{ x = cube2.x , y = cube2.y , z = cube2.z ;
cout << “\nCopy constructor used.” ;}
int volume() {return ( x*y*z);}
}; // end of class
int main()
{
Cubicle cube2(3) ;
{Cubicle cube1;
cout << “Volume of cube1 = ” <<cube1.volume() <<“\n” ;}
cout << “Volume of cube2 = ” <<cube2.volume()<<“\n “;
 Classes and Objects–1 ❖ 281❖
{Cubicle cube3 (4,5);
cout << “Volume of cube3 = ” <<cube3.volume()<<“\n”; }
{ Cubicle cube4;
cube4 = cube2;
cout << “Volume of cube4 = ” << cube4.volume( )<<“\n”; }
return 0 ;
}

The expected output is as below. The output is self explanatory.

Constructor with one parameter used.
Volume of cube1 = 24
Destructor used to remove object.
Volume of cube2 = 18
Constructor with two parameters used.
Volume of cube3 = 60
Destructor used to remove object.
Volume of cube4 = 18
Destructor called to remove object.
Destructor called to remove object.

Also observe that destructor is called when the scope of the object is over.

11.10 ACCESSING PRIVATE FUNCTION MEMBERS OF A CLASS

Carefully see the following program, in which, the data members as well as functions
volume() and surface_area() are declared private. The private functions cannot be accessed
directly from outside. However, they can be accessed through other public member functions
and friend functions of the class. In order that the objects could access private functions, first
we have to introduce public functions which return the values of private functions. The objects
can access these public functions. This is one reason that functions are generally declared public.
But the functions which are used only by class members, and, if it is desired that these should
not be visible to users of the program, may be declared private.

PROGRAM 11.13 – Illustrates that private member functions cannot be accessed

directly.
#include <iostream>
using namespace std;

class Cuboid {

private:
int surface_area( ); // function declared privte
int volume( ); // function declared private
int x , y, z; // data declared private
❖ 282 ❖ Programming with C++

public:
Cuboid(int L,int W,int H ):x(L),y(W),z (H){} // Constructor
}; // End of class
int Cuboid::surface_area()
{return 2*(x*y +y*z +z*x);}
int Cuboid::volume()
{return x*y*z ;}
int main()
{
Cuboid C1(5,6,4);
cout << “ Volume of cuboid C1 ” << C1. volume()<<“\n”;

cout<<“ surface area of C1 = ”<< C1. surface_area()<<“\n”;

return 0 ;
}

The expected output is as follows.

error : cannot access private member declared in class ‘Cuboid’
error : cannot access private member declared in class ‘Cuboid’

The remedy of the above situation is that either to make the functions public in which
case the nature of processing is no longer hidden. However, we can also have public functions
which just pass on the return values of the private functions. In this case the process still
remains hidden, so this is a better option. This method is illustrated in the following program.

PROGRAM 11.14 –Illustrates accessing private function member through a public

member function.

#include <iostream>
using namespace std;
class Cuboid {
private:
int surface_area( ); // private member function
int volume( ); // private member function
int x , y, z; // private data
public:
Cuboid( int L,int W,int H ):x(L),y(W),z (H){} // Constructor
int Surface_area () { return surface_area ();}
//A public function to just pass on the return values.
int Volume () { return volume();}
 Classes and Objects–1 ❖ 283❖
//The public function for passing on the return value
// Volume and volume are different because of different case of
//first letter. So are the Surface_area and surface_area
}; // End of class
int Cuboid::surface_area() // Definition of surface_area()
{return 2*(x*y +y*z +z*x);}
int Cuboid::volume() // Definition of volume
{return x*y*z ;}
int main()
{
Cuboid C1(5,6,4);
cout << “Volume of cuboid C1 ” << C1. Volume()<<“\n”;
cout<<“Surface area of C1 = ”<< C1. Surface_area()<<“\n”;
return 0 ;
}

The expected output is as below.

Volume of cuboid C1 120
Surface area of C1 = 148

The above program illustrates that for accessing a private function one has to have a public
function. In the above example the public function just passes on the value and nothing is
revealed about the function. So it does not dilute the information hiding property of the class.

11.11 LOCAL CLASSES

A class may be declared inside a function, in which case it is called local class. The scope of a
local class is up to the function definition. In the following program two classes are declared
inside a function. The return value of function is calculated from object data of the classes. The
objects are initializes inside the function definition. Such a scheme of programming is not a
popular one.

PROGRAM 11.15 – Illustrates classes local to a function.

#include <iostream>
using namespace std;
int Function ( ) // function definition
{class X // class X declared inside the function
{ private:
int x ;
❖ 284 ❖ Programming with C++

public:
X (int a ) {x = a;}
int getx () { return x ;}
} x_object (5) ; // object of class declared
class Y // Another class declared inside the function
{
private:
int y;
public :
Y(int b ) {y = b;} // constructor function
int gety (){return y ;} // function to access private data
} y_object (10) ; // object definition
// Below is function return value
return (x_object.getx()*(y_object.gety()));
}
void main()
{
cout<<Function ( ) <<endl;
}

The expected output is given below.

50

11.12 STRUCTURES

Structures (a legacy from C language) also give facility to organize related data items in a single
unit. It is a user defined data type containing logically related data but which may be of different
types like int, double, char, arrays or strings. All of them are encapsulated in a unit. For example,
suppose it is required to prepare list of students along with their data like name, registration
number, age, address, etc. This can be done easily by designing a structure of relevant data. Each
student is related to his/her structure, in fact, a structure represents a student. The data items are
called members of the structure. The classes are in fact generalization of structures. A class with
only public members is like a structure. All members of a structure are by default public.
However, in a class declaration, if there is no access specifier, the members are by default private.
But in C++ structures the access specifier as well as functions may be used. This is illustrated
in Program 11.19. Below is an illustration of declaration of a structure like that in C. It starts
with the keyword struct followed by its name and statements are enclosed between curly braces
{ } and a semi-colon after the closing right brace ( }). The declaration does not contain any access
specifier. The structure is made for employees of a company.
struct Employee // Employee is name of struct
{ char Name [30];
 Classes and Objects–1 ❖ 285❖
int Age;
char Designation [25];
int Pay ;
};
Each employee may have an employee-code number such as E1, E2, E3, etc. The code may
be used to represent the structure of a particular employee. In a program the codes E1, E2,
etc., may be defined as an object of type Employee. It is illustrated below.
Employee E1, E2, E3 ;
The above definition allocates enough memory for each of E1, E2 and E3 for storing their
respective data. Note that the declarations of E1, E2, E3 etc., are preceded by the name of
structure Employee because that is their type (user defined type). It is just like we declare an
object of a class. The name of class is the type of its objects. In same way Employee is the type
for E1, E2, E3, etc.

ACCESSING STRUCTURE MEMBERS

The members of a structure are public by default and hence can be accessed by objects or instances
directly. In the above structure, for instance, the relevant data about an employee may be
extracted by using dot (.) operator as it is done for a class object. For instance, if the name,
designation and pay of the employee with the code E3 are to be obtained, it is coded as illustrated
below.
E3.Name ; // The dot ( . ) operator provides the link.
E3.Designation ;
E3.Pay ;

INITIALIZATION
Initialization cannot be done inside the structure declaration because that is a kind of blue-print
or design. The particular instances may be initialized as illustrated below for the case of struct
Employee. The method of accessing the individual structure is illustrated above. If it is desired
to initialize the name of employee for instance, we can write
E1.Name = “ Ram Dass”;
The whole structure may be initialised as illustrated below.
Employee E1, E2, E3;
E1 = {“ Ram Dass”, 30, “Engineer”, 20000};
E2 = {“Dinesh”, 25, “Jr-Engineer”, 15000};
E3 = {“Priti”, 20 , “Office Manager”, 1500};
A structure may be assigned to another structure.

PROGRAM 11.16 – Illustrates a structure.

# include <iostream>
using namespace std;
❖ 286 ❖ Programming with C++
struct Employee // declaration of structure Employee.
{ char Name [30];
int Age;
char Designation [25];
int Pay ;}; // End of structure
int main ()
{ Employee E1 = {“ Ram Dass”, 30, “Engineer”, 20000};
Employee E2 = {“Dinesh”, 25, “Jr-Engineer”, 15000};
Employee E3 = {“Priti”, 20 , “Office Manager”, 1500};
cout<< “The data of employee E2 is :\n”;
cout<< “Pay ” <<E2.Pay<<endl;
cout<< “Name ” << E2.Name<<endl;
cout << “Age ”<< E2.Age <<endl;
cout<< “Designation ” << E2.Designation<<endl;
return 0;
}

The expected output is given below.

The data of employee E2 is :
Pay 15000
Name Dinesh
Age 25
Designation Jr-Engineer
A structure may be assigned to another structure. This is illustrated in the following program.

PROGRAM 11.17 – Illustrates that structures may be assigned.

# include <iostream>
using namespace std;
struct Employee
{ char Name [30];
int Age;
char Designation [25];
int Pay ;}; // end of declaration
int main ()
{ Employee E1 = {“ Ram Dass”, 30, “Engineer”, 20000};
Employee E2 = { “Dinesh”, 25, “Jr-Engineer”, 15000};
Employee E3 = { “Priti”, 20 , “Office Manager”, 15000};
cout<< “The data of employee E2 is :\n”;
cout<< “Pay ” <<E2.Pay<<endl;
cout<< “Name ” << E2.Name<<endl;
cout << “Age ”<< E2.Age <<endl;
 Classes and Objects–1 ❖ 287❖
cout<< “Designation ” << E2.Designation<<endl;
E1 = E3; // assignment of structure
cout<< “Pay ” <<E1.Pay<<endl;
cout<< “Name ” << E1.Name<<endl;
cout << “Age ”<< E1.Age <<endl;
cout<< “Designation ” << E1.Designation<<endl;
return 0;
}

The expected output is given below.

The data of employee E2 is :
Pay 15000
Name Dinesh
Age 25
Designation Jr-Engineer
Pay 15000
Name Priti
Age 20
Designation Office Manager
DECLARATION OF A STRUCTURE IN MAIN()
A structure may be declared inside a function or inside of another structure. The following
program illustrates declaration inside main (). It may contain functions just like classes.
PROGRAM 11.18 – Illustrates declaration of structure in main ().
#include<iostream>
using namespace std;
int main ()
{

struct Employee
{ char Name [30];
int Age;
char Designation [25];
int Pay ;};

Employee E1 = {“Ram Dass”, 30, “Engineer”, 20000};

Employee E2 = {“Dinesh”, 25, “Jr-Engineer”, 15000};
Employee E3 = {“Priti”, 20, “Office Manager”, 1500};
cout<< “The data of employee E2 is :\n”;
cout<< “Pay ” <<E2.Pay<<endl;
cout<< “Name ” << E2.Name<<endl;
cout << “Age ”<< E2.Age <<endl;
cout<< “Designation ” << E2.Designation<<endl;
return 0;
}
❖ 288 ❖ Programming with C++

The expected output is given below.

The data of employee E2 is :
Pay 15000
Name Dinesh
Age 25
Designation Jr-Engineer

C++ STRUCTURES
In C++ structures, the access specifier as well as functions may be used. There is little difference
between a class and a structure. We may use a constructor function as in classes. This is illustrated
in the following program.
PROGRAM 11.19 – Illustrates that access specifiers in C++ structures like in classes.

#include<iostream>
#include <string>
using namespace std;
struct Employee

{
private :
int Pay, Employee_code ;

public:
Employee ( int E,int P ) { Employee_code =E , Pay = P ;}
// constructor function
int getpay ( ) {return Pay ;} // functions
int getcode (){return Employee_code ;}
};

int main ()
{
Employee E1 (22, 10000);
Employee E2 (32, 15000);
cout<< “Employee_code of E1 = ” << E1.getcode ()<<endl;
cout<< “Pay of E1 =” <<E1.getpay()<<endl;
cout<< “Employee_code of E2 = ” << E2.getcode ()<<endl;
cout<< “Pay of E2 =” <<E2.getpay()<<endl;
return 0;
}

The expected output is given below.

Employee_code of E1 = 22
Pay of E1 =10000
Employee_code of E2 = 32
Pay of E2 =15000
 Classes and Objects–1 ❖ 289❖
EXERCISES
1. Give an outline of class declaration.
2. How is an object of a class declared?
3. What makes class program an objected oriented programming?
4. In what way procedural programming is different from object oriented programming?
5. What is the role of a constructor and destructor functions in a class?
6. Can the constructor and destructor functions be declared constants?
7. What is the function of dot (.) operator?
8. What is the difference between the class members declared as private and those declared
public?
9. What is the difference between a private member and a protected member?
10. Why should some members of class be declared as private and others as public?
11. Why member functions of a class are generally declared public and data members as private?
12. Give an example of code for a function prototype declared in class body and function
definition outside the class body.
13. Can you declare an object of a class inside the body of its own class?
14. Can a class definition body include objects of other classes? If so how will you declare the
object in the class?
15. How is, a function declared private in a class, accessed by an object of the same class?
16. How is a function of a class called?
17. What do you understand by following terms.
(i) Information hiding
(ii) Encapsulation of data and functions
18. Make a class program to illustrate the following types of constructors.
(i) Default constructor
(ii) Parametric constructor
(iii) Copy constructor
19. Make a class program with private data members. How can the objects of class access
the same?
20. Make a class program for student registration in which the student is required to enter his/
her name, age and address. The entries should be displayed on monitor so that they can
be verified by the user.
21. Make a class program with a name Polygon for determining the area and periphery of any
regular polygon. Verify the program by determining the area and periphery of following.
(i) Square
(ii) Equilateral triangle
22. Make a class program for determining the area and periphery of ellipses for which the major
and minor axes are declared as private.
❖ 290 ❖ Programming with C++
23. Make a class program which prints the list of grades of a group of 10 students. The list
includes names of students and grades in five subjects.
24. What is the function of destructor function of class?
25. Make a class program with name Rectangle with a constructor function. The class functions
determine the area and circumference of the objects of class. Illustrate the working of class
with an object.
26. What is a structure? How do you define a structure?
27. What are the differences between a class and a structure?
28. Make a program to illustrate an array of structures. Also illustrate input/output for structures.
Answer:
PROGRAM 11.20 – Illustrates an array of structures.

# include <iostream>
using namespace std;

struct Employee

{ char Name [30];

int Age;
char Designation [25];
int Pay ;};
void Display (Employee E )
{
cout<<“Name : ”<<E.Name<< “\t Age ”<<E.Age <<“\n”;
cout<< “Designation ” <<E.Designation<< “ \t\nPay : ”
<<E. Pay<<endl; }

int main ()
{
Employee employee [] = {

{ “ Ram Dass”, 30, “Engineer”, 20000},

{ “Dinesh”, 25, “Jr-Engineer”, 15000},
{ “Priti”, 20 , “Office Manager”, 1500},
{ “ Madhu” , 22, “Office Assistant”, 10000}};

Display (employee [2] );

return 0;
}

The expected output is given below.

Name : Priti Age 20
Designation Office Manager
Pay : 1500

❍❍❍
CHAPTER

12.1 FRIEND FUNCTION TO A CLASS

A friend function to a class has access to all the members of the class whether these are private,
protected or public and this characteristic makes it useful in many applications, particularly, in
overloading of operators. However, the free access to all the members of a class goes against its
frequent use, because, it dilutes the information hiding capability of the class. Since friendship
is granted by the class, therefore, the class declaration should contain the declaration that the
function is friend of the class. Also a friend function is not a member function of the class in
which it is declared as friend, and hence a friend function is defined outside the class body.
The declaration of prototype of a friend function in the class body is illustrated below. It starts
with keyword friend followed by the type of function and identifier (name) for the function
which is followed by a list of types of parameters of the function enclosed in parentheses (). It
ends with a semicolon.
friend type identifier (type_parameter1,type_parameter2, ..);
For instance, a function with name Area() is declared a friend function of class Rect which
is a class for rectangular figures having sides x and y as under.
friend int Area(const Rect &b);
In the above declaration the data of the object b of class Rect are the only parameters of the
function, however, a friend function may have other parameters as well. The definition of the
friend function has to be done outside the class body. The function declared above is defined
below.
int Area (const Rect & b)
{return b.x * b.y ;}
In the above definition the scope resolution operator (::) is not used because Area function
is not a member function of the class.
The following points should be noted about friend functions to classes.
1. If a function F is friend to class B, then F can access all the members of class B.
2. The friendship is granted by the class and not extracted by the function. Therefore, the
friend function has to be so declared in the body of the class as illustrated above.
3. Function prototype of friend function preceded by keyword friend is declared inside the
class body but the function is defined outside the class body.
❖ 292 ❖ Programming with C++
4. Friend function is not a member function of the class to which it is friend. Therefore,
the scope resolution operator is not required while defining the friend function outside
the class body.
5. A class may have any number of friend functions. And a function may be friend to any
number of classes.
6. A friend function may be declared any where in the class. The access specifiers, i.e. public,
protected or private do not affect a friend function.
The following program illustrates the declaration and definition of a friend function.

PROGRAM 12.1 – Illustrates definition of a friend function to a class.

#include <iostream>
using namespace std;
class Rect {
friend int Area(const Rect &a); // friend function Area
int x,y; // private by default
public:
Rect (int L, int W){ x = L,y = W;} // constructor function
}; // end of class
int Area (const Rect &b) // definition of friend function
{return b.x*b.y;}; // scope resolution operator not needed
int main()
{
Rect R1(5,6), R2(3,4) ; //declaration of two objects R1 and R2
cout << “Area of R1= ” << Area ( R1 ) <<“\n”;
cout << “Area of R2 = ” << Area ( R2 )<<“\n”;
return 0 ;
}

The expected output is given below.

Area of R1= 30
Area of R2 = 12
In the above program the friend function takes all the arguments from the data of objects
of the class. One of the object R1 has the dimensions 5 and 6, the friend function gives the area
30 while the second object R2 has the dimensions 3 and 4 and its area is given out as 12. The
output shows that friend function can access the private data members of the class.
A class can have any number of friend functions. In the following program two friend
functions are declared to a class Rect which deals with rectangles. The first friend function
calculates the area taking the private data of the object. The second friend function calculates the
cost of surface treatment of area of rectangle.
 Classes and Objects–2 ❖ 293❖
PROGRAM 12.2 – Illustrates a class with more than one friend functions.

#include <iostream>
using namespace std;
class Rect {
friend int Area(const Rect &a); // a friend function
int x,y; // private by default
public:
Rect (int L,int W){ x = L, y = W;} // constructor function
friend double cost(const Rect &a, double);
//second friend function
}; // end of class
int Area (const Rect &b) // definition of Area Function
{return b.x*b.y;}
double cost(const Rect &b, double s) //definition of cost ()
{return b.x*b.y * s ;}
int main()
{ double A = 4.5 , B = 5.2; // A and B are for data for cost
Rect R1(10,5), R2(20,5) ; // R1, R2 are two objects
cout << “Area of R1= ” << Area (R1) <<“\n”;
cout << “Area of R2 = ” << Area (R2)<<“\n”;
cout << “cost = ” <<cost (R1,A) <<“\n”;
cout << “cost = ” <<cost (R2,B) <<“\n”;
return 0 ;
}

The expected output is given below.

Area of R1= 50
Area of R2 = 100
cost = 225
cost = 520

ONE FUNCTION FRIEND OF MANY CLASSES

A function may be friend to many classes. This is illustrated in the following program in which
a function Display() is declared friend to two classes, i.e. class Sqr and class Rect.

PROGRAM 12.3 – Illustrates a function friend to more than one class.

#include <iostream>
using namespace std;
class Sqr; //pre-declaration of class
❖ 294 ❖ Programming with C++

class Rect { // class Rect

int x,y; // private by default
public:
Rect ( int A, int B) { x = A, y = B;}
int area ( )
{return x*y;}
friend void Display ( Rect R, Sqr S ); // friend function to Rect
};
class Sqr { // class Sqr
int side; // private by default
public:
Sqr (int C){ side = C;}
int Area ( )
{return side*side;}
friend void Display (Rect R , Sqr S ); // friend function to Sqr
}; //end of class
void Display ( Rect R, Sqr S) // Definition of friend function
{cout <<“Area of rect = ” << R.area()<<endl;
cout <<“Area of Square = ” << S.Area()<< endl;}
int main()
{Rect R1(10,5);

Sqr S1 (10);
Display ( R1, S1 );
return 0 ;
}

The expected output is given below. The result is self explanatory.

Area of rect = 50
Area of Square = 100

12.2 FRIEND CLASSES

When all or most of the functions of a class need access to data members and function members
of another class, the whole class may be declared friend to the class. For instance, if the functions
of a class say class A need to access the public, private and protected members of another class
say class B, the class A may be declared as friend class A ; in the body of class B. As you know
friend is a keyword of C++. The declaration is illustrated below.
Class B
{friend class A; //declaration that class A is friend of B
private:
 Classes and Objects–2 ❖ 295❖
statements
friend class C; //declaration that class C is friend of B
public:
Other_statements;
};

CHARACTERISTICS OF FRIEND CLASSES

The following characteristics of friend classes should be noted.
1. If a class A is a friend of class B, then functions of A can access all the members of B.
2. If class A is a friend of class B, it does not mean that class B is also friend of class A. The
C++ friendship is not reciprocal.
3. Friendship is granted by the class whose members are to be accessed. If a class A is
friend of class B, it has to be so declared in the definition of class B.
4. If class A is friend of class B and class B is friend of class C, it does not mean that class A
is also friend of class C or class B is friend of class A or class C is friend of class B or class
A.
The C++ friendship is neither transmitted nor it is reciprocal.
5. A class may be friend to more than one class. And a class may have more than one class
as friends. There is no limit on the number of friends of a class.
In the following program we define a class Cuboid for prismatic components with rectangular
base. The class has two functions to calculate the volume and external surface area of cubicle
elements. A second class with name paint is defined. This class has a function to calculate the
cost of painting the external surface of objects of class Cuboid. The class paint is declared friend
of class Cuboid.

PROGRAM 12.4 – Illustrates declaration of a friend class to a class.

#include <iostream>
using namespace std;
class Cuboid {
friend class paint; // Declaration of friend class
public:
void sides(int , int, int);
int Area();
int volume();
int x , y, z;
}; //end of class Cuboid

void Cuboid:: sides (int L, int W, int H )

{x = L, y = W,z = H; } // Setting the sides of Cuboid

int Cuboid::Area() //Definition of area

{return 2*(x*y +y*z +z*x);}
❖ 296 ❖ Programming with C++
int Cuboid::volume() // definition of volume
{return x*y*z ;}

class paint{ //declaration of friend class paint

private:
int R;
public:
paint () { R = 1;} // default constructor
paint ( int S) { R = S;} // parametric constructor

Cuboid C; // C is the object of class Cuboid

int area (){return C.Area ();}
int cost(int R , Cuboid C ) // R is the rate and C is object
{return R* C.Area () ;} // of Cuboid. cost() is a function
};
int main()
{Cuboid C1 ; // C1 is object of class Cuboid

C1.sides (5,6,5 );
paint P1 ; // P1 is object of class paint
int k = 4;
cout << “Volume of C1 = ” <<C1.volume()<<“\n”;

cout<<“Surface area of C1 = ”<<C1.Area()<<“\n”;

cout<<“Cost of painting P1 = ”<< P1.cost(k, C1)<<“\n”;

return 0 ;
}

The expected output is given below. See carefully the definition and calling of function cost ()
of class paint. P1 is an object of class paint.
Volume of C1 = 150
surface area of C1 = 170
cost of painting P1 = 680

12.3 POINTER TO A CLASS

Pointer to a class may be declared in the same way as it is done for integer or double numbers.
Let us have a class by name List and let L1 be an object of this class. Let ptr be the pointer to
the class. For declaration of pointer to List we may write the code as below.
List L1 ; // declaration that L1 is an object of class List
List *ptr; // declaration of pointer to List
ptr = &(List) L1; // assignment of value to ptr
The last line may as well be written as ptr = &List (L1);
The following program gives an illustration of pointer to class.
 Classes and Objects–2 ❖ 297❖
PROGRAM 12.5 – Illustrates definition of a pointer to a class.

#include <iostream>
using namespace std;
class List
{
private :
int x,y; //let x = the no of items, y = price of one item.
public:

void Setdata (int a, int b ) //A public function to access x,y.

{cout << “Enter number of items : ”; cin >>a ;
cout <<“Enter price of one item : ” ; cin >> b ;
x = a, y = b; }
void Display1 ( )
{cout<<“Number of items = ” <<x <<endl; }

void Display2 ()

{cout<<“Price of each item = ”<<y <<endl ; }

void Display3 ()
{cout<<“Cost of ”<<x<<“ items at the rate of ”<<y<<“per item = ”<<x*y<<endl;}
}; // End of class

void main()
{
List L1 ;
List *ptr; // pointer to List
ptr = &(List) L1; // assigning address of class object

int i,j;
(*ptr).Setdata (i,j); //values to be assigned by user

ptr -> Display1 (); // (*ptr) and ptr-> are equivalent.

ptr -> Display2();

(*ptr).Display3();
}

On running the program, the following text will appear on monitor.

Enter number of items :
Type the data say 50 and enter. Then the text of second line of output will appear on the
monitor, type price say 20 and enter. The following output would appear on the monitor.
Enter number of items : 50
Enter price of one item : 20
Number of items = 50
❖ 298 ❖ Programming with C++
Price of each item = 20
Cost of 50 items at the rate of 20 per item = 1000
The above program also shows that for calling a function of class in the main () through a
pointer ptr, either of the following two codes may be used.
ptr -> Display3();
or
(*ptr).Display3();

Both the above statements are equivalent. In the second statement *ptr has to be enclosed
in brackets because the dot (.) operator has higher precedent than the dereference operator (*).
The following program illustrates, how the functions relating to the second object of the same
class may be called by the use of increment operator (++) on pointer of class pointing to the first
object of the class. Both the objects should be declared before the increment operator.

PROGRAM 12.6 –Illustrates that an increment to pointer to a class points to second

object of class.

#include <iostream>
using namespace std;
class List{
private :
int x,y,z;
public:
void Display1()
{cout<< “Welcome To This Market” << endl; }
void Setprice (int a, int b, int c)
{x = a, y = b, z = c;}
void Display2 ()
{ cout<<“ Price of item1 = ”<< x<<endl;
cout<<“ Price of item2 = ”<< y<<endl;
cout<<“ Price of item3 = ”<< z<<endl; }
}; // End of class
void main()
{ List L1;
List *ptr; //pointer to List
ptr = &(List) L1; // Assignment of address of L1 to pointer
ptr -> Setprice( 6,8,10); // Function relating to L1
ptr -> Display1();
cout << “Price List 1”<<endl;
ptr ->Display2();
 Classes and Objects–2 ❖ 299❖

List L2; // L2 is second object of the class

*ptr++; // Increment of pointer of L1 gives pointer to L2.
ptr -> Setprice(32,27,38); // Function relating to L2.
ptr -> Display1();
cout<< “ Price List 2”<<endl;
ptr -> Display2();
}

The expected output of the program is given below.

Welcome To This Market
Price List 1
Price of item1 = 6
Price of item2 = 8
Price of item3 = 10
Welcome To This Market
Price List 2
Price of item1 = 32
Price of item2 = 27
Price of item3 = 38

In the above program the pointer ptr is declared as pointer to class List with L1 as one of
its object. The functions Setprice() and Display() are called. Obviously these calls are
for L1. Then a second object L2 is declared and ptr is incremented. The incremented value of
ptr points to L2. The functions Setprice() and Display()for L2 are called using this
pointer.

12.4 POINTERS TO OBJECTS OF A CLASS

Let L1 be an object of class list, the pointer to L1 may be declared and assigned as below.
List *ptr = &L1;

PROGRAM 12.7 – Illustrates definition of a pointer to an object of class.

#include <iostream>
using namespace std;
class List{
private :
int x,y,z;
public:
List (int a,int b,int c) {x = a, y = b, z = c;} //Constructor
void Display ()
{cout<<“ Price of item 1 = ”<< x<<endl;
cout<<“ Price of item 2 = ”<< y<<endl;
❖ 300 ❖ Programming with C++
cout<<“ Price of item 3 = ”<< z<<endl; }
}; // End of class definition
void main()
{ int n = 0;
List L1 (12, 15,27) ;
cout << “List ”<<++n <<endl;
List *ptr = &L1; // The (*ptr) and ptr-> are equivalent
ptr -> Display () ;
cout<< “\nList”<< ++n <<“\n”;
List L2 (30, 54,60);
ptr = & L2 ;
(*ptr).Display();
}

The expected output of the program is as below. In this case, the ptr++ does not automatically
point to L2. Therefore, address of L2 is assigned to ptr.
List 1
Price of item 1 = 12
Price of item 2 = 15
Price of item 3 = 27
List2
Price of item 1 = 30
Price of item 2 = 54
Price of item 3 = 60

12.5 POINTERS TO FUNCTION MEMBERS OF A CLASS

A declaration and assignment of pointer to member function of class is illustrated below.

type (class_name ::*pointer_name)( types_parameters) = & class_name :: Function_name;
In the above illustration the first word type denotes the type of return value of the function,
this is followed by, in parentheses, the class identifier, scope resolution operator and indirection
operator (*) and pointer identifier. And in second pair of parentheses, the types of parameters of
the function are listed, each separated by comma. The assignment (the right hand side of =)
comprises address-of operator & followed by class name, scope resolution operator followed by
name of function and it ends with a semicolon. In Program 12.7 given below we have a function
by name Setsides() a public function of class Rect for initialization of sides of rectangle.
It is defined as follows.
void Setsides ( int L, int W ){ x = L ,y = W ;}
The pointer *ptrSet to this function is declared and assigned as below.
void (Rect:: *ptrSet)(int, int) = &Rect :: Setsides;
 Classes and Objects–2 ❖ 301❖
For calling this function for an object R1 with length 20 and width 15 we write the code
as below.
(R1.*ptrSet)(20, 15);
The following program illustrates the use of pointer to a member function of the class.

PROGRAM 12.8 – Illustrates definition of a pointer to member function of class.

#include <iostream>
using namespace std;
class Rect {
int x ,y; // private by default
public:
void Setsides ( int L,int W ){ x = L ,y = W ;}
int Area( ) // Definition of function Area
{return x*y;}
}; // end of class
int main()
{ Rect R1,R2,R3;
void (Rect:: *ptrSet)(int,int) = & Rect :: Setsides;
(R1.*ptrSet) (20,15);
cout << “Area of R1 = ”<< R1.Area()<<endl;
(R2.*ptrSet) (20,30); //object R2 calls function by its pointer
cout << “Area of R2 = ”<< R2.Area()<<endl;
Rect *ptr3 = &R3; // declaring pointer to object R3 of class
(ptr3 ->*ptrSet)(16,10); //calling function by pointer to object
cout << “Area of R3 = ”<< R3.Area()<<endl;
return 0 ;
}

The expected output of this program is as below.

Area of R1 = 300
Area of R2 = 600
Area of R3 = 160

12.6 POINTER TO DATA MEMBER OF A CLASS

The syntax for declaration and assignment of pointer to data members is as follows.
type class_name :: *pointer_name = & class_name :: data_name;
For example, a class with name Rect has data members int x and int y, the pointers ptrx and
ptry to the two data members are declared as below.
❖ 302 ❖ Programming with C++
int Rect :: *ptrx = &Rect :: x;
int Rect :: *ptry = &Rect :: y;
The application is illustrated in a friend function in the following program.

PROGRAM 12.9 – Illustrates definition of pointers to data members and objects of a

class.

#include <iostream>
using namespace std;
class Rect {
friend int Area(Rect a); // friend function
int x, y; // data members private by default
public:
friend double cost(Rect a, double); // another friend function
Rect (int L,int W){ x = L,y = W;} // constructor
}; // end of class
int Area (Rect b) // Definition of area using pointers
{ int Rect :: *ptrx = &Rect :: x; //ptrx is pointer for x
int Rect :: *ptry = &Rect :: y; // ptry is pointer for y
Rect *ptrb = &b; // ptrb is pointer to object b
return b.*ptrx * b.*ptry ;};
double cost(Rect b , double m)
{return b. x* b. y * m ;}
int main()
{ double n = 4.5;
Rect R1(5,6), R2(3,4) ;
cout << “Area of R1= ” << Area (Rect (R1)) <<“\n”;
cout << “Area of R2 = ” << Area (Rect (R2))<<“\n”;
cout << “Cost for R1 = ” <<cost (Rect (R1),n) <<“\n”;
return 0 ;
}

Expected output is given below.

Area of R1= 30
Area of R2 = 12
Cost for R1 = 135

12.7 ACCESSING PRIVATE DATA OF AN OBJECT THROUGH POINTERS

The private data of an object may be accessed through pointers if there is one public data member.
This is possible because the public data members and private data members of an object are stored
 Classes and Objects–2 ❖ 303❖
in sequential blocks of memory. If we know the address of one member data we can determine
the values of other data members by increment or decrement of the pointer. It is illustrated in
the following program. In the program we also find the addresses of the public and private data
by obtaining the values of decremented pointer. However, if you try to determine address of
private data directly by the using &L1.z in the following program the result will be an error
message, “cannot access private member declared in class”.

PROGRAM 12.10 – Illustrates accessing private data of an object through pointers.

#include <iostream>
using namespace std;
class List{
private :
int x,y,z;
public:
int Alpha ;
List (int a, int b, int c ){x = a, y = b, z = c;} //Constructor
}; // End of class definition
void main()
{
List L1( 10, 20, 30);
int *pAlpha ; // pointer to an integer
L1. Alpha = 50;
pAlpha = & L1.Alpha; // assignment of value to pointer
cout<<“Alpha = ”<<*pAlpha<< endl;
cout <<“Address of Alpha = ”<<pAlpha<< endl;
pAlpha––; // decrement of pointer value for getting z
cout <<“The z data of L1 = ”<< *pAlpha <<endl;
cout << “Address of z = ”<< pAlpha<<endl;
pAlpha––; // further decrement of value of pointer for y.
cout << “The y data of L1 = ”<< *pAlpha <<endl;
cout << “Address of y = ”<< pAlpha<<endl;
pAlpha––; // Third decrement to value of pointer for x
cout << “The x data of L1 = ”<< *pAlpha <<endl;
cout << “Address of x = ”<< pAlpha<<endl;
}

The expected output is given below.

Alpha = 50
Address of Alpha = 0012FF7C
The z data of L1 = 30
❖ 304 ❖ Programming with C++
Address of z = 0012FF78
The y data of L1 = 20
Address of y = 0012FF74
The x data of L1 = 10
Address of x = 0012FF70

From the values of addresses you can easily find out that the data members of the class are
stored in the sequential blocks of memory. Each block is 4 byte wide as it is allocated to integers,
this is illustrated in Fig.12.1 below.
x (private) y (private) z (private) Alpha (public)

10 20 30 50

0012FF70 0012FF74 0012FF78 0012FF7C

Fig. 12.1: Addresses of object data
In the following program we start from the address of object. Address of object is also the
address of first data of the object.

PROGRAM 12.11 – Accessing the private data members from address of object.

#include <iostream>
using namespace std;
class List{
private :
int x, y, z;

public:
int s ;

List (int a, int b, int c ) {x = a, y = b, z = c;} //Constructor

}; // End of class

void main()
{
List L1( 10, 20 , 30); ;
L1.s = 40;

int *ptrL1 ;
ptrL1 = (int*)& L1 ;

cout << “Value of x = ”<< *ptrL1<< endl;

cout<<“Address of x = ”<<ptrL1<<endl;
ptrL1++;
cout <<“The y data of L1 = ”<< *ptrL1 <<endl;
cout << “Address of y = ”<< ptrL1<<endl;
 Classes and Objects–2 ❖ 305❖
ptrL1++;
cout << “The z data of L1 = ”<< *ptrL1 <<endl;
cout << “Address of z = ”<< ptrL1<<endl;
ptrL1 ++;
cout << “The s data of L1 = ”<< *ptrL1 <<endl;
cout << “Address of s = ”<< ptrL1<<endl;
}

The value of pointer ptrL1 is the address of first data that is address of x. The increment of
this pointer gives pointer which points to second data that is y. Similarly we get to the third data
that is z. The expected output is given below.
Value of x = 10
Address of x = 0012FF70
The y data of L1 = 20
Address of y = 0012FF74
The z data of L1 = 30
Address of z = 0012FF78
The s data of L1 = 40
Address of s = 0012FF7C

12.8 THE this POINTER

The word this is a keyword of C++ and is the name of pointer to an object of a class. When an
object is created, the compiler creates this pointer which keeps the address of the object. Pointer
this is not a part of object but object has access to it or we can say an object has access to its own
address. When the object calls a member function of the class, this pointer becomes implicit
argument of the function and the function processes the data of the object.
In the following program this as well as *this have been used in the constructor of a class.
Also we find that values of this pointer for two objects C1 and C2 of the class. We also find
the addresses of the objects C1 and C2 by using of address-of operator &. The values of this
pointer for the two objects are equal to their addresses.

PROGRAM 12.12 – Illustrates application of pointer this.

#include <iostream>
using namespace std;
class Cuboid {
public:
Cuboid(int L,int W,int H ){this ->x=L ,(*this).y=W ,this->z=H;}
// the above function is constructor
int surface_area( );
int volume( );
void Display1() // Displays private data of object.
❖ 306 ❖ Programming with C++
{cout <<“x = ”<<this->x <<“, y = ” <<this->y << “, z = ”<<(*this).z <<endl;}

void Display2()
{cout << this<<endl; } // gives value of this
private:
int x ,y, z; // private data
}; // end of class
int Cuboid::surface_area() // definition of surface area
{return 2*(x*y +y*z +z*x);}

int Cuboid::volume() // definition of volume

{return x*y*z ;}

int main()
{ Cuboid C1(5,6,4), C2(7,8,5) ; // C1 and C2 are two objects
C1.Display1();
C1.Display2(); // value of this pointer for C1

cout <<&C1<<endl; // Address of C1

C2. Display1();
C2. Display2(); // value of this for C2
cout << &C2<<endl; // Address of object C2

cout << “Volume of cuboid C1 = ”<<C1.volume()<<“\n”;

cout<< “Volume of cuboid C2 = ”<< C2.volume()<<“\n”;
cout<<“Surface area of C1 = ”<< C1.surface_area()<<“\n”;
cout<<“Surface area of C2 = ”<< C2.surface_area()<<“\n”;
return 0 ;
}

The expected output is given below.

x = 5, y = 6, z = 4
0012FF74
0012FF74
x = 7, y = 8, z = 5
0012FF68
0012FF68
Volume of cuboid C1 =120
Volume of cuboid C2 = 280
Surface area of C1 = 148
Surface area of C2 = 262
 Classes and Objects–2 ❖ 307❖
From the output of the program it is clear that value of this pointer and address of respective
objects are equal. Also the statement (*this).z has been used to determine the value of z data
of object. The this pointer points to the object of class.

12.9 STATIC DATA MEMBERS OF A CLASS

Objects of a class keep their own copy of data members of class. In certain situations a part of
data may be same for all objects of a class. In such cases instead of every object keeping an
individual copy of this data, it would be prudent if one copy of the data is shared by all the objects
of class, because, this would save memory space, particularly, when the common data is large.
For such applications the common data may be declared static. The static data may be declared
under any one of the three access specifiers, i.e. public, protected or private. A public static data
member may be accessed directly through any object of the class because it is same for all the
objects. The code comprises the object name followed by dot operator followed by name of data
member. Static data member declared private or protected may be accessed through the
corresponding public functions of the class.
For declaration of static data the usual declaration of a variable is preceded by the keyword
static. This is illustrated in the following program.

PROGRAM 12.13 – Illustrates declaration of static members in a class.

#include <iostream>
using namespace std;
class Cuboid {
public:
static int x; // x declared static
Cuboid( int W ): y(W) {} // constructor
static void Display ( ) // static function
{cout <<“ Height of all objects is = “ << z <<endl;}
int surface_area( );
int volume( );
private:
int y;
static int z ; // static variable
}; // end of class
int Cuboid :: x = 10;
int Cuboid :: z = 5;
int Cuboid::surface_area()
{return 2*(x*y +y*z +z*x);}
int Cuboid::volume()
{return x*y*z ;}
❖ 308 ❖ Programming with C++
int main()
{
Cuboid C1(6), C2(3) ;
cout << “C1.x = ”<< C1.x << “, C2.x = ”<<C2.x <<endl;

Cuboid :: Display( ); // function call without an object

cout << “Volume of cuboid C1 ” << C1.volume()<<“\n”;
cout<< “Volume of cuboid C2 = ” << C2.volume()<<“\n”;
cout<<“surface area of C1 = ”<< C1.surface_area()<<“\n”;
cout<<“surface area of C2 = ”<< C2.surface_area()<<“\n”;
return 0 ;
}

The expected output is as below.

C1.x = 10, C2.x = 10
Height of all objects is = 5
Volume of cuboid C1 300
Volume of cuboid C2 = 150
surface area of C1 = 280
surface area of C2 = 190

In the above program two data members x and z are declared static. The int x is declared static
under public domain while the int z is declared static under access label private. The data x ( a
public member) may be accessed directly by an object of the class. See the output C1.x =10 and
C2.x = 10, in the first line of output. The static data declared private cannot be accessed directly
by an object. It has to be accessed through a public function like any other private data member.
In the above program it is illustrated by function Display () which displays value of z.
In case of fundamental type static data members, they are by default initialized to 0 at the
time of declaration. Putting any other value in the declaration may give error. Static members
are defined at file scope. The x and z in the above program are defined outside the main () in
file scope as below.
int Cuboid :: x = 10;
int Cuboid :: z = 5;

12.10 STATIC FUNCTION MEMBER OF A CLASS

A function member of a class may also be declared static provided its arguments are also declared
static or if it does not access any non-static member of the class. The pointer this associated
with every object of class is applicable only to non-static function members of the class. The static
function members of class do not have this pointer. The following program illustrates the
application of a static function with static parameters.
 Classes and Objects–2 ❖ 309❖
PROGRAM 12.14 – Illustrates static member function of a class.

#include <iostream>
using namespace std;

class Cuboid {

public:
Cuboid(int H):z(H) {} //constructor
static int Base_area (){ return x*y;} // static function

int surface_area( );
int volume( );

private:
static int x , y; // x and y declared static
int z ;
}; // end of class

int Cuboid :: x = 5; // static data member

int Cuboid :: y =8 ; // static data member

int Cuboid::surface_area() // definition of surface_area ()

{return 2*(x*y +y*z +z*x);}

int Cuboid::volume() // definition of function volume()

{return x*y*z ;}
int main()
{
Cuboid C1(5), C2(8) ;
cout<<“Base area of all objects = ”<< Cuboid::Base_area()<<endl;

cout << “Volume of cuboid C1 ” << C1.volume()<<“\n”;

cout<< “Volume of cuboid C2 = ” << C2.volume()<<“\n”;

cout<<“Surface area of C1 = ”<< C1.surface_area()<<“\n”;

cout<<“Surface area of C2 = ”<< C2.surface_area()<<“\n”;

return 0 ;
}

The expected output is given below.

Base area of all objects = 40
Volume of cuboid C1 200
Volume of cuboid C2 = 320
Surface area of C1 = 210
Surface area of C2 = 288
❖ 310 ❖ Programming with C++

12.11 DYNAMIC MEMORY MANAGEMENT FOR CLASS OBJECTS

OPERATORS NEW AND DELETE AND NEW[] AND DELETE[]

The operator new allocates memory from the free store called heap while the operator delete
does just the opposite, it releases the memory by deleting the object. The released memory is
added to heap for other uses. For allocation of memory for an array we have the operator new[]
and corresponding operator delete[] for removing the array and releasing the memory. For
fundamental types we have already discussed these operators in Chapter 9. For class objects new
may be used for allocating memory for single object while new[] may be used for creating an
array of class objects. Similarly single object is removed from memory by operator delete and
array of objects are removed by delete[] operator. It must be noted that when new [] is used
for allocation of memory, for releasing the memory we must use delete [] and not simply delete.
For example if you have created an array of objects by new[] and if you use simply delete to
remove the array, it will only delete the first object of the array, the remaining objects would not
get deleted. So it will be only partial release of memory. The following program illustrates the
application of new and delete for class objects.

PROGRAM 12.15 – Illustrates application of operators new and delete for class objects.
#include <iostream>
using namespace std;

class List {
private :
double x ; // x = weight in kg, y = price of one kg.
double y;

public:
void Setdata (double a, double b )
{x = a; y = b; }

void Display ( )
{cout<<“ Weight = ”; cin >> x;
cout<<“ Price = ”; cin >>y;
cout<<“ Cost of ”<<x<<“kg at the rate of ”<<y<<“ per kg = ”<<x*y<<endl;}
};
void main()
{double i,j;
List *ptr= new List ; // use of new operator
(*ptr).Setdata (i,j);
ptr -> Display (); // (*ptr) and ptr-> are equivalent.
delete ptr;
}

The expected output is given below. By now the reader should be able to follow the
implemention of the program.
 Classes and Objects–2 ❖ 311❖
Weight = 8.45
Price = 10.0
Cost of 8.45kg at the rate of 10 per kg = 84.5

The following program illustrates the application of new[] and delete[].

PROGRAM 12.16 – Illustrates the operators new[] and delete[] for an array of objects.

#include <iostream>
using namespace std;
class List {
private :
int x,y; // x = the no of items, y = price of one item.
public:
void Setdata (int a, int b )
{x = a;
y = b;
}
void Display ( )
{cout<<“ Number of items = ”; cin >> x;
cout<<“ Price of each item = ” ; cin >>y;
cout<<“ Cost of ”<<x<<“ items at the rate of ”<<y<<“ per item = ”<<x*y<<endl;}
}; // End of class
void main()
{
List *ptr= new List[3]; //pointer to List
for (int k=0; k<3; k++)
{cout<< “ For item Number ” <<k+1<<endl;
int i=0,j=0; //for user input of data
(*ptr).Setdata (i,j);
ptr -> Display ();} // (*ptr) and ptr-> are equivalent.
delete[]ptr;
}

The expected output of the above program is given below.

For item Number 1
Number of items = 20
Price of each item = 50
Cost of 20 items at the rate of 50 per item = 1000
For item Number 2
❖ 312 ❖ Programming with C++
Number of items = 30
Price of each item = 10
Cost of 30 items at the rate of 10 per item = 300
For item Number 3
Number of items = 40
Price of each item = 60
Cost of 40 items at the rate of 60 per item = 2400
In the above programs, we have been having the class and the main program in continuity.
The class programs may be stored in a separate file and included in the program by the header
#include “class_name” (see Appendix E for more details.)
Besides, the operator new may be used in the constructor function itself so that memory is
allocated at run time. The following program illustrates a class program for matrices.

12.12 A MATRIX CLASS

The following class deals with the input/output of matrices. The functions such as addition,
subtraction or multiplication are included in the main program. However, another class may be
developed for carrying out these operations and the following class may be inherited.
PROGRAM 12.17 – Illustrates addition, subtraction and product of matrices.
# include <iostream>
using namespace std;
class Matrix
{
private:
int m, n;
public :
int** Ptr;

Matrix ( int R , int C )

{ m = R, n = C ; // R stands for row and C stands for column

Ptr = new int *[m];

for( int i =0; i<m; i++)
Ptr[i] = new int [n];}
// E_Val stands element value
void Read_Elements ( )
{ cout <<“Enter the elements of ”<<m <<“x”<<n << “ matrix: ”;
int E_Val =0;
for( int r =0; r< m; r++)
for ( int k =0; k<n;k++)
{ cin >> E_Val;
Ptr [r][k] = E_Val; } }
void Write_Elements ( )
{
 Classes and Objects–2 ❖ 313❖
for(int s =0; s<m; s++)
{for ( int p =0; p<n ; p++)
cout<< Ptr[s][p]<<“\t ”;
cout<<endl;}}
};
int main()
{
Matrix A (3,3), B (3,3), C(3, 3), D(3,3), E(3,3);
cout <<“Enter elements of matrices A and B. \n”;
A.Read_Elements( );
B.Read_Elements ( );
cout <<“For matrix A you have entered the elements as below.\n”;
A.Write_Elements( );
cout <<“For matrix B you have entered the elements as below.\n”;
B.Write_Elements( );
cout << “A.Ptr[1][0] = ” << A.Ptr[1][0]<< endl;
for ( int k = 0; k<3; k++ )
for ( int p =0; p<3; p++)
{C.Ptr[k][p] = A.Ptr[k][p] + B.Ptr[k][p]; // Matrix addition
D.Ptr[k][p] = A.Ptr[k][p] – B.Ptr[k][p];} //Matrix Subtraction
for(int J =0;J<3 ; J++) // Multiplication
for (int K =0; K<3;K++)
{ E.Ptr[J][K] =0;
for ( int S =0 ; S<3 ; S++)
E.Ptr[J][K] += A.Ptr[J][S] * B.Ptr[S][K];}
// Below are output statements
cout<< “Matrix C = Matrix A + Matrix B :”<<endl;
cout << “The elements of matrix C are as below ”<<endl;
C.Write_Elements();
cout<< “For Matrix D = Matrix A – Matrix B :”<<endl;
cout << “The elements of matrix D are as below ”<<endl;
D.Write_Elements();
cout<< “For Matrix E = Matrix A * Matrix B :”<<endl;
cout << “The elements of matrix E are as below ”<<endl;
E.Write_Elements();
return 0;
}

The expected output is given below. The output is quite obvious.

Enter elements of matrices A and B.
❖ 314 ❖ Programming with C++
Enter the elements of 3x3 matrix: 1 2 3 4 5 6 7 8 9
Enter the elements of 3x3 matrix: 10 20 30 40 50 60 70 80 90
For matrix A you have entered the elements as below.
1 2 3
4 5 6
7 8 9
For matrix B you have entered the elements as below.
10 20 30
40 50 60
70 80 90
A.Ptr[1][0] = 4
Matrix C = Matrix A + Matrix B :
The elements of matrix C are as below
11 22 33
44 55 66
77 88 99
For Matrix D = Matrix A – Matrix B :
The elements of matrix D are as below
–9 –18 –27
–36 –45 –54
–63 –72 –81
For Matrix E = Matrix A * Matrix B :
The elements of matrix E are as below
300 360 420
660 810 960
1020 1260 1500

12.13 LINKED LISTS

A linked list is a data structure in which the items are linked together through pointers. A linked
list consists of nodes. In case of singly linked lists, each node contains the data items which may
be fundamental type or class/structure object and a pointer to the next node. Such a list can be
traversed only in the forward direction. In a doubly link lists, each node contains the data item,
a pointer to next node and a pointer to previous node. Such a list can be traversed in both
directions, i.e. forward as well as backward. Figures.12.2 (a and b) illustrate the node structures
of singly linked lists and doubly linked lists respectively.

Data Pointer to next node

Fig. 12.2(a): A node of singly linked list

 Classes and Objects–2 ❖ 315❖

Pointer to previous node Data Pointer to next node

Fig. 12.2(b): A node of doubly linked list

The nodes may be stored anywhere in the memory. In case of arrays the data items are stored
in a linear way, in adjacent memory locations and the elements are linked because of their location.
If the address of one array elements gets known addresses of all the elements can be determined.
In case of linked lists, however, the items are not stored in adjacent memory blocks but a node
knows the address of the next node but not of next to next. Therefore the traversal is possible
only from one node to the next node. Random access is not possible. Figure 12.3 illustrates a
singly linked list and Figure 12.4 illustrates a doubly linked list.

First (&N1)

Node N1 Node N2 Node N3

John Mitra SippiMira Jyoti Kara

(021) 2665 3421 (021)2665 3421 (451) 2665 3426

Next(&N2) Next(&N3) Next(&N4)

Node N4 Node N5
Sultan Amd Sukla Gour
(011) 2665 3423 (041) 2665 3490

Next(&N5) Next(NULL )

Fig. 12.3: Singly linked list—The data item is in shaded rectangle and pointer in the other

Fig. 12.4: Doubly linked list—Data items are in shaded rectangle, pointers in other two
❖ 316 ❖ Programming with C++
The following two programs which are developed from ideas of illustrations in references
[1, 2, 5 and 8] illustrate a singly linked list and a doubly linked list respectively. In Program
12.18, a list of students roll numbers and grades is made. The entries done at random are inserts
in appropriate places in the list so that it makes a sorted list. In Program 12.19 a doubly linked
list of persons is made, the data item is the telephone numbers.

PROGRAM 12.18 – Illustrates singly linked list

# include <iostream>
#include < string>
using namespace std;
class Student
{
int Roll_Number; // private by default
int Grades ;
public:
Student() {Grades = 0;}
~ Student(){}; // destructor function, below is constructor
Student ( int N , int Marks) { Roll_Number = N, Grades = Marks;}
int getRN ( ) { return Roll_Number;} // for accessing private
int getGrad () { return Grades ;} // data
};
class Node
{
Student * Pstudent; //pointer to student
Node * Pnext; // pointer to next node
public:
Node (Student *);
~Node () {} ;
void setnode ( Node * node) {Pnext = node;}
// public function to set Pnext
Student * getStudent() const{return Pstudent;} // for access
void Display_Grades(void)
{ if ( Pstudent ->getGrad() >= 0) // display if grades >=0
cout <<“ ” << (*Pstudent).getRN() <<“ \t \t ”<< Pstudent -> getGrad()<<“\n”;
if ( Pnext)
Pnext-> Display_Grades();}
void Insert ( Node* new_node) // function for inserting new_node
{ if (Pnext ==0 ) // if it is first
Pnext = new_node;
//if it is not the first determine, place to insert new_node
 Classes and Objects–2 ❖ 317❖
// by comparing values of previous node, new_node and next node
else
{ int next_Grades = Pnext->getStudent() ->getGrad();
int new_Grades = new_node->getStudent()->getGrad();
int Current_Grades = Pstudent->getGrad ();
if ( Current_Grades < new_Grades && next_Grades >new_Grades )
// if new_node is between current_node and next_node
{new_node ->setnode( Pnext); // new_node = Pnext and
Pnext = new_node;} // Pnext = new_node
else
// if above two options are not there, it is last node
Pnext -> Insert(new_node);}
}
};
Node :: Node ( Student * Pst ):Pstudent ( Pst),Pnext (0){}
int main()
{
int Marks;
int rollnumber;
Node * pnode =0;
Student *Pstudent = new Student(0,0);
Node *Start= new Node(Pstudent); // defines the start of list
while ( true)
{cout<<“Enter rollnumber : ”; cin >> rollnumber;
if ( rollnumber == 0)
// condition for getting out of endless loop
break;
cout<< “Enter Marks :”; cin >> Marks; // get data Marks
Pstudent = new Student( rollnumber,Marks); // new student points
pnode = new Node(Pstudent); // new node
Start -> Insert(pnode );} // insert new node
cout << “Roll-Number Marks” <<endl; // top line of display
Start -> Display_Grades(); //Display values
delete Start;
return 0;
}

An example of its application is given below.

Enter rollnumber : 21 // Entry of data
Enter Marks :76
Enter rollnumber : 23
❖ 318 ❖ Programming with C++
Enter Marks :57
Enter rollnumber : 24
Enter Marks :87
Enter rollnumber : 32
Enter Marks :95
Enter rollnumber : 33
Enter Marks :65
Enter rollnumber : 0
Roll-Number Marks //output of program
0 0
23 57
33 65
21 76
24 87
32 95
The various comments given in the program help explain the output. It places the nodes such
that it makes a sorted list as illustrated in the output.
The following program illustrates doubly linked list using structures. Each node contains the
address to the next node as well as to the previous node. It is also illustrated in Fig. 12.5. After
entries of names and numbers, a search is made for the number of a certain person. Two trial
searches are made one for person in the list and one for person not in the list. See the output.

PROGRAM 12.19 – Illustrates doubly linked list by using structures.

# include <iostream>
#include <string>
using namespace std;
struct Node
{string Name ;
string Tnumber ;
Node * Next; // pointer to next node
Node* Prev ; // pointer to previous node
}; // end of structure

void Display1(Node *next)

{while (next != NULL )
{ cout << next –> Name<<“ \t ”<<next –>Tnumber<<endl;
next = next–>Prev;}}

void Search(Node *S , string Str)

{while ( S != NULL)
if ( S->Name == Str )
{cout << “Result of search is;\n”;
cout<< S->Name <<“ \t ”<< S->Tnumber<<endl;
break;}
 Classes and Objects–2 ❖ 319❖
else S = S->Next;
if( S == NULL)
{ cout << “Result of search is;\n”;
cout<<“The name ”<< Str<< “is not in the list\n”;}}

int main()
{ Node N1, N2,N3,N4,N5;
N1.Name = “John Mitra” ;
N1.Tnumber = “(021) 2665 3421” ;
N2.Name = “SippiMira”;
N2.Tnumber = “(013) 2665 3434”;

N3.Name= “Jyoti Kara”;

N4.Name= “Sultan Amd”;
N5.Name= “Sukla Gour” ;
N3.Tnumber = “(451) 2665 3426”;
N4.Tnumber = “(011) 2665 3423”;
N5.Tnumber =“(041) 2665 3490" ;

Node *Last =&N5;

N5.Prev = &N4;
N4.Prev = &N3;
N3.Prev = &N2;
N2.Prev = &N1;
N1.Prev = NULL;

Node *First = &N1;

N1.Next = &N2;
N2.Next = &N3;
N3.Next = &N4;
N4.Next = &N5;
N5.Next = NULL;

Display1 (Last);
Search(First, “SippiMira”);
Search ( First, “Gurmeet”);
return 0; }

The entries and expected outputs are given below.

Sukla Gour (041) 2665 3490
Sultan Amd (011) 2665 3423
Jyoti Kara (451) 2665 3426
SippiMira (013) 2665 3434
John Mitra (021) 2665 3421
Result of search is;
SippiMira (013) 2665 3434
Result of search is;
The name Gurmeet is not in the list
❖ 320 ❖ Programming with C++

12.14 NESTED CLASSES

A class may be declared inside another class. The access specifiers work on nested class in the same
way as they act on other functions. In the following illustrative program class Z is declared in
class Y in the public domain and class Y is in turn is declared in class X in public domain. All
the three classes have private data for which appropriate constructor functions are provided in
each class. The objects of each of these are declared in the main() and values assigned to them.

PROGRAM 12.20 – Illustrated nested classes.

#include <iostream>
using namespace std;
class X
{private:
int x ;

public:
X (int a ) {x = a;}
int getx () { return x ;}

class Y // nested class Y declared in class X

{private:
int y;
public :
Y( int b ) {y = b;}
int gety (){return y ;}

class Z // class Z declared in class y

{ private:
int z ;
public :
Z (int c ) { z = c;}

int getz () { return z;}

}; // end of class Z
}; // end of class Y
}; // end of class X
int main()
{ X x_object (5);
X::Y y_object (10);
X::Y::Z z_object (20);
cout<<“x_object = “<<x_object.getx() <<endl;
cout << “y _object = “<< y_object.gety()<<endl;
cout<< “z_object = “ <<z_object.getz()<<endl;
return 0 ;
}

The expected output is given below.

x_object = 5
y _object = 10
z_object = 20
 Classes and Objects–2 ❖ 321❖
EXERCISES
1. What is a friend function to a class?
2. What are the privileges of a friend function with respect to the class?
3. How is a friend function declared and defined?
4. How to declare a friend class?
5. What are characteristics of a friend class?
6. What is a static data member of a class?
7. What is a static function member of a class? Write an illustrative code for declaration of a
static function member of a class.
8. What is the difference between the static data member of class and constant data member
of a class?
9. How do you declare and initialize a static data member of a class?
10. What is this pointer? How is it related to the object of class?
11. Give an example of code in which this pointer is explicitly used.
12. Give an example of code which declares a pointer to member function of a class.
13. How would you declare a pointer to object of a class?
14. How do you declare pointer to data member of a class? Write an illustrative code.
15. Which function of class is called when an object of a class is created and which function is
called to destroy it?
16. What for we use the pointer new and pointer new[ ]? Write the corresponding destructors.
17. What do you understand by ‘dynamic memory management’. Which operators/functions are
used for the purpose?
18. Make a class program to illustrate declaration of pointer to the object of class.
19. Make a class program for regular polygons with number of sides and length of side as private
data. Declare two friend functions one for calculation of area of polygon and the other for
circumference of polygon.
20. Make a class program for ellipses with major axis and minor axis as private data. Make a friend
class to this program to calculate the circumference and area of ellipses.
21. Make a class program and illustrate the declaration and use of pointer to a member of class.
22. Make a class program to illustrate the declaration and use of pointer to the class.
23. Make a program to register the applicant names and their addresses. Make use of pointer new
to register the next applicant. The program should also display the information on the monitor.
24. Fill in the blanks with the appropriate words given in brackets.
(a) A friend function is defined ——— of class. (inside or outside)
(b) A class C is declared as friend of class B in the body of class B. With this declaration
class B ———— friend of class C. (becomes, does not become).
(c) The pointer —— enables to call multiple functions in the same statement. (this or new )
(d) A friend function to a class has access to ——— members of a class. (only public
members, only private members, all members)

❍❍❍
❖ 322 ❖ Programming with C++

CHAPTER

13.1 INTRODUCTION

The operators such as + , –, * etc. are used to carry out certain operations and are frequently used
in program statements. C++ has a rich store of such operators which are defined for the
fundamental types such as int, float, double, char variables. For example let us consider the
following statements.
int z = int x + int y ;
bool( y == z );
In execution of first statement the value of int x is added to the value of int y and the result
is assigned to int z. The actions of operator + and operator = are already defined in C++. The
operations are automatic in the sense that the programmer does not have to write any additional
code except to write the above statements in a program and run it. Similarly in the second
statement the value of y is compared with the value of z and if they are equal the outcome is
true or 1 and if they are not equal the output is 0. The mode of comparison is already defined
for fundamental types. In the above case each of objects x, y and z has only one value at a time.
For user defined types such as class objects in which case an object may comprise several items
of data, how can you compare whether one class object is equal to another class object or assign
the values connected with one object to another object of the class. For example let us consider
a class graph and let Graph1 and Graph2 be the two objects of the class (Figure 13.1a). Now
if it is desired to compare the two graphs in order to find if Graph1 is equal to Graph2 how
do we write code for this comparison. Figure 13.1b shows another example – two vectors V1
and V2 as objects of a class Vector. As you know a vector has magnitude as well as direction.
Now it is desired to compare the two vectors to check if the two are equal or add the two vectors
or assign a vector to another vector, just like we often do for int, float and char variables. Also
C++ does not allow invention of new operators. We have to use the operators that we use for
the fundamental types. Straight away we cannot write V1 = V1 as we write for int variables unless
we redefine how the operator = is going to work in this particular case. C++ does allow us to
redefine the functionality of these operators so that they can perform the operations as we expect
them to do for objects of a particular class. The redefinition of the functionality of operators
for class objects is called operator overloading. It is not only elegant to use the same operators
 Operator Overloading ❖ 323❖

to do similar operations with class objects, it also makes the class objects on par with the
fundamental types as far as operators are concerned.
In case of graph (Fig. 13.1a) we may code that if the lengths of lines in their respective order
are equal and the lines are similarly placed in the two graphs then the two graphs are equal. For
instance, starting from left if the line on the extreme left of Graph1 is equal to the line on the
extreme left of Graph2, and similarly if all other lines are equal in the two graphs and they are
similarly placed we can take the two graphs as equal. For dealing with vectors, in algebra we take
components of vectors along chosen directions represented by axes of a coordinate system.
Addition of two vectors involves addition of respective components, i.e. add x-component of
vector V1 to x-component of vector V2 and repeat the same for other components. The
comparison of two vectors also involves the comparison of respective components. For example,
the two vectors are equal, if x-component of one is equal to x-component of the other and
y-component of one is equal to y-component of the other and so on. In C++ the number of
vector elements or components are not limited by the number of axes of co-ordinates. Like arrays
there may be any number of elements in a vector. However, the same algebraic concepts are
applied for operations like assignment, addition, comparison etc. For example the two vectors
are equal if their elements in the respective positions are equal.

V1
V2

Graph 1 Graph 2
(a) (b)

Fig. 13.1

The functionality of all the operators of C++ except a few may be redefined for user defined
types (class objects) to carry out the operations as user wants them to do. This is called
overloading of operators. Remember operator overloading is applicable only to user defined
types, or to a mix of user defined types and the fundamental types. For the fundamental types
the operations of different operators are already defined in C++ Standard Library and these cannot
be altered by operator overloading.
The overloading of operators is a very important aspect of C++ and is highly useful in object
oriented programming. It gives us the facility to assign one class object to another, or compare
two class objects and carry out several conditional statements using the same operator symbols
as they are used for fundamental types. Thus operator overloading gives a vast extension to C++.
The overloading of operators, however, is not an exclusive property of C++. Other object oriented
computer languages also provide this facility. For instance, the concept of operator overloading
was first used in Fortran, though then the scope was very much limited. C++ provides a much
more general tool which is highly important in OOP. In this chapter we shall discuss the operators
that may be overloaded, the methods to carry it out and the over-riding conditions.
❖ 324 ❖ Programming with C++

13.2 OPERATORS THAT MAY BE OVERLOADED

Most of the operators of C++ except a few may be overloaded. Table 13.1 gives the list of
operators which can be overloaded and there are only 5 operators given below the table which
cannot be overloaded.

Table 13.1 – Operators that can be overloaded

+ – * / % = & | ~ ^

== > < != <= >= += –= *= /=

%= &= ^= |= >> << >>= <<= && ||

! unary++ Unary – – Unary + unary – , [] () –> –>*

new new[] delete delete[]

Among the operators listed in the table above, the following operators may be overloaded
both in their unary and binary form.
+ – * &

THE OPERATORS THAT CANNOT BE OVERLOADED

The following operators cannot be overloaded.
. .* :: ?: sizeof

RESTRICTIONS ON OVERLOADED OPERATORS

The overloading of operators is subject to following restrictions or conditions.
1. Operators can be overloaded only for user defined types or a mix of user defined types
and fundamental types.
2. The number of operands that an operator takes (arity) cannot be changed.
3. The operator precedence cannot be changed in overloading.
4. The operator associativity (left to right or right to left) cannot be changed in over-
loading.
5. There is no restriction on defining the functionality (behaviour) of overloaded operator.
However, it is desirable and advisable that an overloaded operator should carry out similar
operation on class objects as it does for the fundamental types. For instance, if operator + does
addition for fundamental types, it should do same for class objects.

13.3 OPERATOR OVERLOADING FUNCTIONS

For overloading of an operator, we define a function which redefines the functionality of the
operator being overloaded, i.e, the overloading function defines what the operator will do and
how it will do that operation when it is applied to objects of a particular class. The operator
 Operator Overloading ❖ 325❖

overloading function is a non-static function (because it has to operate on non-static arguments).

The overloading function is declared with first word type (type of data it returns) followed by
the keyword operator which is followed by the operator symbol which is desired to be overloaded
followed by argument in parentheses (see the declaration below). The operator overloading
function may be a class function member, it may be a friend function of class or it may be a
global function. The choice of function is discussed below. For example for overloading operator
(+) we may have it as class member function. In class body the overloading function is defined
in the following way.
type operator + (arguments)
{ function_ body }
Here type refers to the type of data it returns. If it returns a class object then name of class is the
type. Should the function be a class function, a friend function or a global function? Both the
class function and friend function have access to all the members of a class, i.e. private, protected
and public. A global function cannot have direct access to private and protected members of a
class, therefore, additional public functions will have to be included for accessing private and
protected members of class. This impairs the information hiding capability of class as well as it
is an additional load on the performance. Therefore, either a class function or friend function
should be used as far as possible. For overloading the function call operator, i.e. (), subscript
operator, i.e. [], indirection operator i.e. -> and assignment operator, i.e. = , the operator
overloading function should be a class member.

OVERLOADING FUNCTIONS AND ARGUMENTS

In case of friend functions and global functions a unary operator overloading function will have
one argument and a binary operator overloading function will have two arguments. However,
for class member functions, the class object is the argument which is implicitly available to the
operator overloading function. Therefore, for overloading functions which are class members only
one argument is required for binary operators and none is required for unary operators. For
selecting between the class member function and global/ friend function the following guideline
may be used.
(a) In the application of the operator, if the leftmost operand is an object of the class for
which the overloading function is being defined, the function may be a class member.
However, friend functions or global functions may also be used.
For example, let V1 and V2 be the two objects of a class Vector, for operators like = we
write V1 = V2 ; In this code, the left most operand VI is a class object therefore,
overloading function may be a class member function.
(b) If the leftmost operand is not an object of the class for which the overloading function
is being defined then a global or a friend function is needed [2]. For example, suppose
we want to redefine the functionality of operators << and >> for vectors having more
than one values. For output/input we wrote the code cout << class object or cin>> class
object. Here cout and cin are not objects of the class for which the overloading functions
are to be defined. For such cases friend functions or global functions may be used.
Considering the performance, friend functions are superior to global functions, hence
friend functions are preferred.
In the following program the operator + is overloaded to perform addition of two vectors.
❖ 326 ❖ Programming with C++

PROGRAM 13.1 – Illustrates overloading of operator + for addition of two vectors.

#include <iostream>
using namespace std;
class Vector{
private:
int x, y, z ; // three components of a vector
public :
Vector(int a,int b,int c){ x =a ; y = b; z =c;} //constructor
Vector (){}; // Empty or default constructor
Vector operator+ (const Vector&S) //for overloading binary operator +
{ //only one argument needed.
x =x +S.x;
y = y + S.y;
z = z + S.z; //The leftmost operand is class object
return Vector (x,y,z);} // so class member function is used.

void Display() {
cout << “x component = ”<< x << endl;
cout << “y component = ”<< y << endl;
cout << “z component = ”<< z << endl;}
}; // End of class
int main()
{ Vector V1 ( 2,4,5), V2( 6, 5, 8), V3;
V3 = V1 +V2;
V3.Display( ) ;
return 0;
}

The expected output is given below. The overloaded operator adds the respective components
of the two vectors, i.e. 2+6 =8, 4+5 = 9 and 5 + 8 = 13.
x component = 8
y component = 9
z component = 13
If the operator overloading is defined outside the class, the function prototype must be
declared inside the class body. For the above case, in the class body we write function prototype
as given below.
Vector operator + (Vector);
The function definition placed outside the class would be coded as below.
Vector Vector :: operator+ (const Vector&S)

x = x +S.x;
y = y + S.y;
z = z + S.z;
return Vector (x,y,z); }
This is illustrated in Program 13.2 below. The last line in the above could also be written as
return *this ;
 Operator Overloading ❖ 327❖

OPERATOR + DEFINED TO CARRY OUT MINUS OPERATION

In overloading, the functionality of the operator is redefined, so it is possible that + operator may
be defined to carry out minus (see Program 13.2) or multiplication or division (/). But it is a
not a good practice. It is advised that the native functionality of operator should not be changed
in overloading. Because, after some time the programmer may forget the redefinition of the
operator and it can lead to serious errors in use of the program as well in debugging the program.
However, it is for the sake of illustration that following program is included.

PROGRAM 13.2 – The operator + is overloaded to carry out minus operation.

#include <iostream>
using namespace std;
class Number {
private:
int x, y ;
public :
Number (int a, int b ){ x = a ; y = b;} // constructor
Number(){}; // empty or default constructor
Number operator+(const Number & m); // function prototype
void Display()
{cout << “The x component = ”<< x << endl;
cout << “The y component = ”<< y << endl; }
} ; // End of class
Number Number :: operator+ (const Number & m) // function definition
{ // outside the class
x =x – m.x;
y =y – m.y;
return Number (x,y); }
int main()
{
Number N( 16, 21 ), M( 5,6 ), D, E; //N, M, D, E are objects
D = N + M;
D.Display( ) ;
int A = 10; // A and B are fundamental type integers
int B =5;
cout << “A + B = ” << A+B <<endl;
return 0;
}

The expected output is given below:

The x component = 11
The y component = 15
A + B = 15
❖ 328 ❖ Programming with C++

The output shows that on class objects the operator + has carried out minus operation, i.e.
x =16–5=11, in the first line and 21–6 =15 in the second line of output. However, the native
functionality (addition) of operator + for fundamental types is not lost by overloading which
is for a class objects only. This is shown in third line of output by addition of two integers A
and B in which case it is doing addition while for class objects it is doing minus operation. The
program also illustrates how to define the overloading function outside the class.

13.4 ADDITION OF COMPLEX NUMBERS

A similar case is that of addition of complex numbers. In algebra we write a complex number
as x + y i. Here x is the real part, and yi where i = √–1 is the imaginary part. Suppose we want
to add two complex numbers such as (6 + 5i) + (4 + 8i). The result in algebra is 10 + 13i, that
is the real part is added to real part and imaginary part is added to imaginary part. The operator
+ in C++ with its native functionality defined for fundamental types cannot do such an addition.
However, C++ allows us to overload the operator + such that it does the same type of addition
as it is carried out in algebra for complex numbers.

PROGRAM 13.3 – Illustrates addition of two complex numbers.

#include <iostream>
using namespace std;

class Number
{
int x, y ; //complex number is x + yi, private by default

public :
Number (int a, int b ) { x =a ; y = b;} // constructor
Number (){}; // empty or default constructor

Number& operator+(const Number & m) //redefinition of operator +

{
x = x + m.x;
y = y + m.y;
return* this; }

void Display(){
cout<<“The resultant number is = “<< x <<“ + “<< y <<“i”<< endl;}
} ; // End of class

int main()
{ Number N( 16, 21 ), M( 5,6 ), D ;
D = N + M;
D.Display( ) ;
return 0;
}
 Operator Overloading ❖ 329❖

The expected output is given below. The program is similar to that for vector additions.
The resultant number is = 21 + 27i

13.5 OVERLOADING OF += AND – = OPERATORS

In the following program combination operators += and –= are overloaded.

PROGRAM 13.4 – Illustrates overloading of += and – = operators.
#include <iostream>
using namespace std;
class Vector
{
private:
int x, y, z ; // three components of vector

public :
void setvalues (int a, int b, int c) { x =a ; y = b; z =c;}

void operator += (Vector P) // Definition of overloading

{ x = x + P.x ; // function
y = y + P.y ;
z = z + P.z ; }

void operator –= ( Vector P) // definition of overloading

{x = x – P.x ; // function
y = y – P.y ;
z = z – P.z ; }

void Display( )
{ cout << “x = “<< x <<“\ty = “<<y << “\tz = “<< z <<endl;}
}; // End of class

int main()
{ Vector V1, V2, V3;
V1.setvalues ( 12,4,6);
V2.setvalues( 2, 5, 7);
V3.setvalues (4,5,6) ;
V1 += V2;
V3 –= V2;
cout<<“ Components of V1 are : “;
V1.Display() ;
cout<<“ Components of V3 are : “;
V3.Display();
cout<<“ Components of V2 are : “;
V2.Display();
return 0;
}
❖ 330 ❖ Programming with C++

The expected output is given below.

Components of V1 are : x = 14 y=9 z = 13
Components of V3 are : x = 2 y=0 z = –1
Components of V2 are : x = 2 y=5 z=7

The components of V1 are obtained by adding components of V2 to components of V1

and assigning the sum to components of V1, so in the output we get component of V1 as 14,
9 and 13. In case of V3 the components of V2 are subtracted from the components of V3 and
result is assigned to components of V3. So, we get components of V3 as 2, 0 and –1.

13.6 OVERLOADING OF INSERTION (<<), EXTRACTION ( >>) AND /=

OPERATORS

Since the left side operands of << and >> are not the objects of class for which the overloading
function is being defined, therefore, we have to use either a global function or a friend function.
The friend function can access all the members of a class, therefore, in the following program
we use a friend functions for overloading << and >> operators. Program also illustrates
overloading of /= operator for an array.

PROGRAM 13.5 – Overloading of insertion (<<), extraction (>>) and /= operators.

#include <iostream.h>
const int n = 5;
class List {
friend ostream & operator << ( ostream & , List & ) ;
friend istream & operator >> ( istream & , List & ) ;
float x[n] ; // private by default
public :
List(){}; // empty or default constructor
List (float a[n]) // constructor
{for ( int i =0; i<5; i++)
x[i] = a[i];}
float operator /=(int m) // overloading of /= operator
{ for ( int i =0; i<5; i++)
x[i] = x[i]/m;
return x[i];}
}; // End of class
istream & operator>> ( istream &In, List&L) // Friend
{ for ( int i = 0 ; i< n ; i++) // function definition
In >> L.x[i]; // In is an object of istream so type is
return In;} //istream. L is an object of List
ostream & operator<< ( ostream &Put, List &L ) //friend
{Put<<“(”<<L.x[0]; //function definition.
 Operator Overloading ❖ 331❖

for ( int i = 1; i<n; i++) //Put is defined an object of ostream

Put<<“, ”<< L.x[i] ; // so type is ostream
Put<<“)”;
return Put ;}
int main()
{List L1 , L2;
float B[n] ={2.2,3.3,4.4,5.5,6.6};
cout<<“Put in ”<<n<<“ elements of L2: ”;
cin>> L2;
L1 = B ;
L1 /=2;
cout<<L1<<“\n”;
L2 /=3;
cout<<L2 <<endl;
return 0;}

The expected output is as below. The output is self explanatory.

Put in 5 elements of L2: 3.6 6.6 9.9 12.66 15.99
(1.1, 1.65, 2.2, 2.75, 3.3)
(1.2, 2.2, 3.3, 4.22, 5.33)

13.7 OVERLOADING OF INCREMENT AND DECREMENT OPERATORS

(++ AND – –)

The unary operator ++ can be prefixed before a name of variable or class object or postfixed after
the name of variable or object. In the first case, it first increments the value of variable or object and
then the value is used for evaluation of following expression. In postfix the initial value is used for
evaluation of expression and then the increment is carried out. The unary operator – – behaves in a
similar fashion with decrement of values.
PROGRAM 13.6 – Illustrates overloading of ++ and – –

#include <iostream>
using namespace std;
class Vector
{
private:
int x, y, z ; // three components of vector
public :

Vector(int a, int b,int c) { x =a ; y = b; z =c;}

Vector (){}
Vector operator ++(int) // Postincrement operator ++
❖ 332 ❖ Programming with C++

{ x +=1;
y += 1;
z +=1;
return Vector( (x-1), (y-1), (z–1));}

Vector operator ++() // Preincrement operator++

{ x +=1;
y += 1;
z +=1;
return Vector (x,y,z);}
Vector operator –– () // Predecrement operator––
{ x –=1;
y –= 1;
z –=1;
return Vector (x,y,z);}
Vector operator ––(int) // Postdecrement operator ––
{ x –=1;
y –= 1;
z –=1;
return Vector( (x+1), (y+1), (z+1));}

void Display( )
{ cout << “x = “<< x <<“, y = “<<y <<“, z = “<< z <<endl;}

}; // End of class B

int main()

{ Vector V1 (4,5,6), V2(21,22,23), V3 ( 1,2,3), V4(11, 12, 13) ;

cout<< “Initial V1-components are: ”;

V1.Display();

cout<< “After post increment V1-components are: ”;

(V1++).Display();
cout<<“Now V1 components are : ” ;
V1.Display();

cout<< “Initial V2-components are: ” ;

V2.Display();

cout<< “After preincrement V2-components are: ”;

(++V2).Display () ;

cout<< “Initial V3-components are: ”;

V3.Display();
 Operator Overloading ❖ 333❖

cout<< “After post decrement V3-components are : ” ;

(V3––).Display();
cout<<“Now components of V3 are : ” ;
V3.Display();

cout<< “Initial V4-Components are :” ;

V4.Display();

cout<< “After predecrement V4-components are :”;

(––V4).Display () ;
return 0;
}
The expected output is given below. The three components of the vector V1 are 4, 5 and 6 and
after post increment these values are same. However, the vector is incrementd as is clear from
the next line of output. Same is true for post decrement for vector V3. See the 6th, 7th and 8th
lines of output. The initial components 21, 22, 23 of V2 become 22, 23 and 24 after pre-
increment. In case of predecrement of vector V4 its components 11, 12 and 13 become 10, 11
and 12 after the operation of overloaded pre-decrement. The detailed output is given below.
Initial V1-components are: x = 4, y = 5, z = 6
After postincrement V1-components are: x = 4, y = 5, z = 6
Now V1 components are : x = 5, y = 6, z = 7
Initial V2-components are: x = 21, y = 22, z = 23
After preincrement V2-components are: x = 22, y = 23, z = 24
Initial V3-components are: x = 1, y = 2, z = 3
After postdecrement V3-components are : x = 1, y = 2, z = 3
Now components of V3 are : x = 0, y = 1, z = 2
Initial V4-Components are :x = 11, y = 12, z = 13
After predecrement V4-components are :x = 10, y = 11, z = 12

In the above program the actions of preincrement and predecrement as well as postincrement
and postdecrement for class objects are illustrated.

13.8 DOT PRODUCT OF VECTORS

In vector mathematics the dot product involves multiplication of corresponding components

(components with same index value) of two vectors and adding the results of multiplications.
The following program illustrates it.

PROGRAM 13.7 –Overloading (*) using friend function in order to carry out the inner
product of two vectors.

#include <iostream>
using namespace std;
class Vector{
❖ 334 ❖ Programming with C++

private:
int x, y, z ; // three components of vector
public :
Vector (int a, int b, int c) { x =a ; y = b; z =c;}
Vector (){};
friend int &operator*(const Vector &, const Vector & ); // two parameters
}; // End of class
int & operator* (const Vector &S, const Vector &P ) // function definition
{ int Innerproduct;
Innerproduct= S.x * P.x + S.y * P.y + S.z *P.z ;
return Innerproduct;}
int main()
{ Vector V1 ( 3,5,5);
Vector V2( 10,6,8);
int dotproduct = V1 * V2;
cout << “dotproduct of V1 and V2 = “<<dotproduct<<endl;
return 0;
}

The expected output is as below.

dotproduct of V1 and V2 = 100

The output of the above program gives the dot product of two vectors each having three
components. The program can be easily extended for n components. The following program
overloads operator == for two vectors with class member function.

13.9 OVERLOADING OF EQUALITY OPERATOR (==)

When we are dealing with quantities like vectors, complex numbers, etc., the comparisons
whether two vectors are equal, involves comparison of each component of vector V1 with the
corresponding component of vector V2. The two vectors are equal if the components of one
vector are equal to the corresponding components of the other vector. So we have to redefine
the functionality of == operator. This is illustrated in the following program.

PROGRAM 13.8 – Illustrates overloading of operator ( == ) for vectors.

#include <iostream>
using namespace std;
class Vector
{
int x, y, z ; // private by default
public :
Vector (){x = 3, y = 2, z =1;} // default constructor
 Operator Overloading ❖ 335❖

Vector (int a, int b,int c) { x =a ; y = b; z =c;}

void operator==( Vector S)

{if ( x==S.x && y == S.y && z== S.z)
cout << “true”<< endl ;

else
cout<<“false”<< endl; }

void Display ()
{
cout << “Vector = (“<< x << ”, “ <<y << ” “ <<z<<”)”<< endl;}
} ; // End of class

int main()
{
Vector V1 ( 2,4,5), V2( 6, 5, 8) ,V3 ( 6,5,8);
V1 == V2;
V2 == V3;
V1.Display ();
V2.Display();
V3.Display ();
return 0;
}

The expected output is given below.

false
true
Vector = (2, 4, 5)
Vector = (6, 5, 8)
Vector = (6, 5, 8)

13.10 OVERLOADING OF INDEX OPERATOR []

In C++ there is no provision to check whether the index or subscript value of an array element
called is within the number of elements of the array or not. If it is greater than the number of
elements in the array, a garbage value may return leading to errors. The following program, by
overloading [] operator provides a check for this and gives a message if the subscript value is out
of bounds.

PROGRAM 13.9 – Illustrates overloading of [ ] operator.

#include <iostream.h>
#include <cstdlib>
❖ 336 ❖ Programming with C++

const int n = 5;
class List {
friend ostream & operator << ( ostream & , List & ) ;
int x[n] ; // private by default
public :
void Setvalues (int a[n] )
{for ( int j = 0; j< n; j++)
x[j] = a[j] ; }
int operator [](int k)
{ if (k <0 || k > n)

{ cout << “Array index k is out of bound”;

exit (0);}
return x[k];}
}; // End of class List
ostream & operator<< (ostream &Put , List &L )
{Put<<“The components are :(”<<L.x[0];
for ( int i = 1; i<n; i++)
Put<<“, ”<< L.x[i] ;
Put<<“)”;
return (Put) ;}
int main()
{
List L1 ;
int B1[] = {11, 12, 13, 14,15};
L1.Setvalues (B1);
cout<< L1 <<endl;
cout << L1[2] << endl;

cout<< “The following result is for L1[7] ”<<endl;

cout << L1[7]<<endl;
return 0;
}

The expected output is given below.

The components are : (11, 12, 13, 14, 15)
13
The following result is for L1[7]
Array index k is out of bound

The output given above is self explanatory. The first line of output is a normal output of
an array. In the second line the third component of array has been called. This is 13. Remember
 Operator Overloading ❖ 337❖

that the subscripts start from zero. Next the eighth component has been asked while the array
contains only 5 elements. So the overloaded subscript operator gives the message that the index
is out of bound and terminates the program.

EXERCISES
1. What is operator overloading?
2. What is the need for operator overloading?
3. Which operators cannot be overloaded?
4. How do you define an overloading function?
5. How do you decide whether overloading function should be a class function, a friend function
or global function?
6. How many arguments a class function would take for overloading following operators?
(i) + (binary) (ii) +=
(iii) – unary (iv) ==
7. What kind of function (class, friend or global) may be used for overloading following operators?
(i) + operator (ii) >> operator
(iii) += operator (iv) ++ or –– operator
8. How many arguments friend functions would take for overloading following operators.
(i) << operator (ii) ++ operator
(iii) *= operator (iv) = = operator
9. Can the operator overloading be used for fundamental type data?
10. Which of the following operators can not be overloaded?
(i) == (ii) .
(iii) :: (iv) .*
(v) /=
11. What are the conditions (or restrictions) on overloaded operators?
12. Give instances of practical class objects and operations for which the following overloaded
operators may be used.
(i) >> and << (ii) ==
(iii) ()
13. Which type of function (class member or friend function) be used for overloading the following
operators?
(i) << (ii) []
(iii) >> (iv) –>
(v) /= (vi) ==
14. Make a program to illustrate the overloading of operators << and >>.
15. Make a program for overloading operator (*) for multiplication of two complex numbers.
16. Make a program for overloading operator (+) for adding two strings.
17. Make a program for overloading operators new and delete.
❖ 338 ❖ Programming with C++

18. Make a program for overloading new [ ] and delete [ ].

19. Make a program to illustrate the overloading of || (OR) and && (AND) operators.
20. Make a program to illustrate the inner product of two vectors each having 10 components.
Answer:

PROGRAM 13.10 - Inner product of two10 dimension vectors.

#include <iostream>
using namespace std;
const n =10 ;
class Vector
{
int x [n] ; // n components of vector/ array
public :
Vector (int A [n]) {for (int i = 0 ; i < n ; i++)
x[i] = A[i] ;}
Vector (){};
friend int operator* ( Vector P , Vector S );
} P,S ; // End of class
int operator* ( Vector P , Vector S )
{int IP =0;
for ( int j =0; j< n; j++)
IP = IP + P.x[j] * S.x[j] ;
return IP;}
int main()
{
int C1[ ] = { 4,3,2,3,2,5,3,1,2,3 };
int C2[ ] = {4,2,1,3,2,1,3,1,4,2};
Vector V1 (C1);
Vector V2(C2);
int dotproduct = V1 * V2 ;
cout << “dotproduct of V1 and V2 = “<<dotproduct<<endl;
return 0;
}

The expected output of the program is given below.

dotproduct of V1 and V2 = 66

21. Change the Program 13.4 by changing the function void operator +=() to Vector operator +=(),
Make other required changes.

❍❍❍
CHAPTER

14.1 INTRODUCTION

Inheritance is an important feature of software reuse and object oriented programming (OOP).
Suppose you have already made a class program for describing some features of a group of
objects. And you also know that a subgroup of them possesses some special features over the
others. Instead of making a complete class program starting from scratch to describe the special
features along with the ones for which you have already made the class program, you can better
make a new class program which inherits the existing class as it is with all its functions and data
members and adds the additional features that you wish to describe for the special group of
objects. You are thus reusing the existing class program which has already been debugged and
tested without doing any modification to it. The new class is called derived class and the existing
class is called base class. The process does not end here, you may have another derived class
which inherits the features of the derived class or features of the derived class as well as features
of some other classes. When a class inherits a single class, it is called single inheritance. This is
illustrated in Fig.14.1. Say class D is the derived class and class B is the base class, in case of
single inheritance, the derived class is declared as below.
class D : access_specifier B

Base class class B

Access_specifier (public or protected or private)

Derived class class D

Fig. 14.1: Single inheritance, example of declaration (class D: public B)

The access specifier is either public or protected or private. For example for public
inheritance the code for declaration of class D would be
class D : public B
C++ also supports inheritance from more than one class in many different ways as illustrated
in Fig.14.2 below. The classes from which a class is derived may or may not be related to one
another.
❖ 340 ❖ Programming with C++

In real life also, objects derive several of their characteristics from the characteristics of other
objects. Thus a garment is as good as the fabric from which it is made. Similarly, the characteristics
of wooden furniture are related to the type of wood it is made from. The following figures
illustrate some more examples of inheritance in real life objects.

Polygon

Bolts

Hexagon Quadrilateral Triangle

Equilateral triangle
Square
Hexagonal Rectangle
Headed- Bolts

Table Top

Fig. 14.2: The figure shows the shapes and their connection with physical objects

Figure14.2 shows the complex way in which the polygonal shapes are related to one another
and to industrial products. Even the design of industrial products like cars, etc., are influenced
by customer’s perceptions (Fig. 14.3). This shows that real life products are interconnected in
a complex way.
class Customer_perceptions

Engine Body Accessories

class Car_designs

Fig. 14.3

As explained above, inheritance is an important feature of real life objects. It is also a very
important feature of C++ because it is instrumental in reusing the existing software and in
building really big programs. Thus a big program may be subdivided into suitable base class,
derived classes and objects.

14.2 FORMS OF INHERITANCES

The base classes and derived classes may be related in a number of ways as illustrated in the
Fig.14.4 (a – e). Codes for declaration of derived classes are also given below each figure.
 Inheritance ❖ 341❖

class Wood Base class class Fabric Base class

class Furniture class Men_Garment class Lady_Garments

Example of declaration Example of declaration of derived classes

class Furniture: public Wood class Men_Garment: public Fabric
class Lady_Garment: public Fabric

Fig. 14.4 (a): Single inheritance Fig. 14.4 (b): Hierarchical inheritance

class A
class B1 class B2 class B3

class B

class C
class D

Example of declaration
class A
{ — }; Example of declaration

class B : public A class D: public B1,public B2,private B3

{---};
class C: public B
{---} ;

Fig. 14.4 (c): Multilevel inheritance Fig.14.4 (d): Multiple inheritance

class Director

Admininstration
class Admininstration class Faculty

class Student

Example of declaration of class Student

class Student : public Faculty, public Admin, public Director

Fig.14.4 (e): Multi-path inheritance

❖ 342 ❖ Programming with C++

As illustrated in the above figures, the inheritance may be declared public, private or
protected. However, in all the declarations of derived classes the first word is class – a keyword.
This is followed by name of derived class, followed by colon (:) and then followed by access
specifier and name of base class. In case of more than one base class, the access specifier has to
be mentioned individually with each class. Let class B be the base class and class D be the derived
class. The declarations of class D in case of single inheritances may be coded as below.
class D : public B // public inheritance
class D : protected B // protected inheritance
class D : private B // private inheritance
class D : B //No access specifier, private by default

Figures 14.4(a– e) also show the codes for declaration of derived class in several different types
of inheritances.

14.3 SINGLE PUBLIC INHERITANCE

In this kind of inheritance the public functions of base class may be accessed directly by derived
class members, i.e. derived class inherits these functions. The objects of derived class can directly
access the public functions of base class. Out of the private and protected members of base
class, only protected members may be accessed by derived class through its own public function
members while the private members of base class are not visible in derived class. The private
members of base class can be accessed through public and protected function members of base
class. However, friend functions of base class also have access to all the members (public, private,
protected) of base class but friend functions are not inherited. Objects of derived class are
also objects of base class but objects of base class are not the objects of derived class.
If the base class has only public members and inheritance condition is also public then all
the members of the base class can be directly accessed by member functions as well as by objects
of derived class. Programs 14.1 to 14.3 illustrate the public inheritance with base class having
the following categories of members.
(i) Base class has only public members, (Program 14.1).
(ii) Base class has public and protected members, (Program 14.2).
(iii) Base class has public and private members, (Program 14.3).
PROGRAM 14.1– Illustrates public inheritance with the base class having only public
members.

#include <iostream>
using namespace std;
class B
{
public :
int m ,a ;
int Product1 () { return a*m;}
}; // End of base class B
 Inheritance ❖ 343❖
class D : public B // Declaration of derived class D
{
public:
int n ;
int Product2 () { return n* Product1();}
}; // End of class D

int main()
{ D C1; // C1 is an object of class D
C1.m = 5; // data directly accessed by object
C1.a = 4 ;
C1.n = 3 ;
cout<< “Product1 = ”<<C1. Product1 ()<< endl;
cout << “Product2 = ”<< C1.Product2()<<endl;
return 0;
}

The expected output is given below.

Product1 = 20
Product2 = 60
In the above program, the object C1 of derived class D has direct access to the public
functions of base class. This is illustrated by the statement C1.m = 5; and the statement
C1.Product1(),etc.
If the base class has public as well as protected members, all the members of base class can
be accessed by members of derived class. The protected members of base class cannot be directly
accessed by objects of derived class, however, they may be accessed though public member
functions of the derived class.

PROGRAM 14.2– Illustrates public inheritance with base class having only public and
protected members.

#include <iostream>
using namespace std;

class B // declaration of class B

{
protected :
int m ; // protected member of class B
public:
int k;
int Square (){return k*k ;}
int Product () { return k*m;}
}; // End of class B
❖ 344 ❖ Programming with C++
class D : public B //class D inherited from class B
{
public:
void setvalue1(int a) // public function to access m of B
{m = a;}
}; // End of class D.
int main()
{ D C; // C is declared an object of D
C.k = 9; //Direct access to k of class B
C.setvalue1(6); //Access to m through public member of D
cout<< “Square = ” << C.Square()<< “, Product = ”<< C.Product() <<endl;
return 0;
}

The expected output is given below.

Square = 81, Product = 54
In the above program the protected data member of class B has been accessed via a public
function member of class D, i.e. void setvalue1(int a){m = a ;}
The private members of base class cannot be accessed by public functions of derived class.
A private member of base class when accessed by members of derived class would result in an
error signal as illustrated in the output of the following program.

PROGRAM 14.3 – Illustrates public inheritance with base class having private data.

#include <iostream>
using namespace std;
class B
{
private : // private access
int m ;
}; // End of class B
class D : public B // Declaration of class D
{
public:
int a;
void setvalue( ) // Function for accessing m of class B
{m = a;}
int n ;
}; // End of class D
int main()
{ D C; // C is an object of class D
 Inheritance ❖ 345❖
C.a = 5 ;
C.n = 4 ;
int product= C.a* C. n;
cout<< “ Product of members of B and D = ”<< product<< endl;
cout << “ Sum of squares of members of B & D = ”<< C.n*C.n + C.a*C.a<<endl;
return 0;
}

The expected output is an error message of the type given below.

cannot access private member declared in class B
If it is required to access the private data members or private function members of base class,
corresponding public functions which provide access to private members have to be included in
base class. These functions provide the interface. These are like brokers in trade. The private
members of a base class can only be accessed through public and protected functions of base
class or by friend functions of base class. But friend functions are not inherited.
However, protected member of base class may be accessed by public member function of
derived class. This has already been illustrated in Program 14.2. Also see the following program.

PROGRAM 14.4 –Illustrates public inheritance when base class has private data members
and function members.

#include <iostream>
using namespace std;
class B { // base class B
private :
int x; // private member of B
int Square (){ return x*x ;} // private member of class B
public:
void setvalue (int p ); //setvalue()-a public member for
// accessing the private member x of B
int Psquare(){return Square ();} // A public member of B to
//access private member Square()
int m ;
}; // End of class B
void B::setvalue (int p){x = p;}; // Definition of setvalue()
class D : public B //declaration of D with public inheritance
{
public:
int n ;
}; // End of class D
int main()
❖ 346 ❖ Programming with C++
{ D C; // C is an object of class D
C.setvalue(5) ; // accessing x of B through setvalue ()
C.m = 4 ; // accessing public member of B directly.
C.n = 3 ; // accessing public member of D directly.
cout<< “Product of m and n = ” << C.m*C.n<< endl;
cout<< “Square = ” << C.Psquare() << endl;
return 0;
}

The expected output of the program is given below.

Product of m and n = 12
Square = 25
In class B of the above program there is a private data member int x; and public data
member int m; and a private function member int square(){return x*x ;}. The
private data of class B is accessed by defining a public function void setvalue (int p);
in class B and the int square() is accessed by declaring another public function in class B,
i.e.
int Psquare(){return Square ();}
and this public function is accessed directly by objects of derived class D. Also note that we cannot
write as below.
C.m*C.setvalue(5)
This is because the function setvalue() is declared void (no return value) and hence
use of arithmetic operators is not legal with void functions.

14.4 SINGLE PROTECTED INHERITANCE

In case of protected inheritance the public members and the protected members of base class
become protected members of derived class. The private members of base class are not visible
to derived class. They can only be accessed through public and protected member functions of
base class. The following program illustrates the protected inheritance with base class having
protected and public members.

PROGRAM 14.5 – Illustrates protected inheritance when base class has protected and
public data as well as function members.

#include <iostream>
using namespace std;
class B {
protected : // Access protected
int m;
int Square () {return m*m;}
public : // Access public
int k ;
 Inheritance ❖ 347❖
product () {return k*m;}
}; // End of class B
class D : protected B // Inheritance protected
{ public:
void setvalue1(int a ) // public member to access m
{m = a;}
void setvalue2( int p) { k = p;} // public member to access k
int dsquare( ){ return B:: Square () ;} //For accessing Square()
int dproduct(){ return B:: product();} //For accessing product()
}; // End of class D, C is an object
int main()
{ D C; // C is an object of D
C.setvalue2(10);
C.setvalue1(5);
cout<<“Product of m and k = ” << C.dproduct() <<endl;
cout<<“Square = ” << C.dsquare( ) <<endl;
return 0;
}

The expected output of the above program is given below.

Product of m and k = 50
Square = 25
In case of protected inheritance the public and protected members of base class become
protected members of the derived class. These may be accessed by public member functions of
derived class.
But the private members cannot be accessed by members of the derived class directly. They
can only be accessed through public and protected member functions of base class itself. In the
following program we have a base class with private data members as well as private function
member. For accessing these two members we have to create public functions in the base class
itself. For private data m we have the public function set_value() and for private function
Square() we have the public function bSquare() which returns the return value of
Square().

PROGRAM 14.6 – Protected inheritance with base class having private members.

#include <iostream>
using namespace std;
class B { // base class
private :
int m; // private member data
int Square(){return m*m ;} // private member function
❖ 348 ❖ Programming with C++
public :
void setvalueb(int a ) {m = a;} // public member to access m
int bSquare(){return Square ();} // for accessing Square
int k ; //public member of B
int product (){return k*k;}
}; // End of class B
class D : protected B // protected inheritance
{
public:
void setvalued (int n) { setvalueb(n);}
void setvalue2(int p) { k = p;} // For accessing k,
int dbsquare(){return bSquare ();} // For accessing bSquare,
int dproduct(){return product ();} //for accessing product
}; // End of class D.
int main()
{ D C; // C is an object of class D
C.setvalue2(10);
C.setvalued(5) ;
cout<<“Product = ” << C.dproduct() <<endl;
cout<<“Square = ” << C.dbsquare( ) <<endl;
return 0;
}

The expected output is given below. The output is self explanatory.

Product = 100
Square = 25

14.5 SINGLE PRIVATE INHERITANCE

With private inheritance the public and protected members of base class become private members
of derived class. Hence they cannot be accessed directly by an object of derived class. However,
they can be accessed through public function members of the derived class. The private members
of base class are not visible in the derived class, they can be accessed only through public and
protected member functions of the base class.

PROGRAM 14.7 – Illustrates that in private inheritance public and protected members of
base class become private members in derived class.

#include <iostream>
using namespace std;
class B {
protected :
int m ;
 Inheritance ❖ 349❖

public:
int k;
}; // End of class B
class D : private B // private inheritance
{
public: // access specifier in class D
int a;
void setvalue( ) {m = a;}
int n ;
};
// End of class D
int main()
{DC; // C is an object of class D
C.k = 6;
C.a = 5 ;
C.n = 4 ;

int product= C.a* C. n;

cout<< “ Product of members of B and D = ”<< product<< endl;
cout << “ Sum of squares of members of B & D = ”<< C.n*C.n + C.a*C.a<<endl;
return 0;
}

The expected output is an error message that k cannot be accessed directly.

‘K’; cannot access public member declared in class B
In the following program the inheritance condition is private. The base class has one protected
data member m, one public data member k and a public function Square (). For the derived class
all these become private. Therefore, public functions are developed in the derived class so that
these can be accessed by the objects of derived class.

PROGRAM 14.8 – In private inheritance the public and protected members of base class
are accessed through public functions of derived class.

#include <iostream>
using namespace std;
class B{
protected :
int m;
public :
int k;
int Square () {return k*k;}
❖ 350 ❖ Programming with C++
int msquare () {return m*m;}
}; // End of class B

class D : private B // private inheritance

{
public:
int a;
void setvalue1() // public member of D to access
{m = a;} //protected member of B
void setvalue2 (int b) { k = b;}

// public member of D required to access public member of B.

//Because of private inheritance. Same applies to following also.
int dmsquare () {return msquare () ;}
int Dsquare(){ return B:: Square () ;}
}; // End of class D

int main()
{ D C; // C is an object of class D
C.setvalue2(6);
C.a = 5 ;
cout<<“ Square = ” << C.Dsquare() <<endl;
cout <<“square of m = “<<C.dmsquare ()<<endl;
return 0;
}

The expected output is given below.

Square = 36
Square of m = 25
From the results of above programs the access condition in the three types of inheritances
have been summarised below in Table 14.1. Here D is the name of derived class and B is the
name of base class.
Table 14.1 (A) – Public inheritance (class D: public B)

Base class access specifier Access in the derived class

public Public.Directly accessible by member functions and objects of
class D and its friend functions.
protected Protected. These can be accessed by member functions of
derived class and its friend functions.
private Not accessible directly by members in derived class. Private
members of base class can be accessed, only through public and
protected member functions of base class, by member functions
and friend functions of derived class.
 Inheritance ❖ 351❖
Table 14.1 (B) – Protected inheritance (class D: protected B)

Base class access specifier Access in the derived class

public Protected. These can be accessed only by member functions
and friend functions of derived class.
protected Protected. These can be accessed only by member functions
and friend functions of derived class.
private Not visible in derived class. These can be accessed only through
public and protected members of base class.

Table 14.1(C) – Private inheritance (class D: private B)

Base class access specifier Access in the derived class

public Private. These can be accessed by member functions and friend
functions of derived class.
protected Private. These can be accessed by member functions and friend
functions of derived class.
private Not visible in derived class. These can be accessed only through
public and protected member functions of base class.

From the above table it is clear that class members which are desired to be inherited should
be declared either public or protected in the base class.

14.6 MULTIPLE INHERITANCE

The following figure illustrates the multiple-inheritance in which a class is derived from several
base classes.

Base classes class B1 class B2 class B3

Derived class class D

Fig. 14.5: Multiple-inheritance—Class D derived from three base classes B1, B2, B3

In the following program the class D is derived from two base classes B1 and B2. The rules
of inheritance have already been discussed above. The same rules apply to multiple inheritances
as well.
❖ 352 ❖ Programming with C++

PROGRAM 14.9 – Illustrates multiple-inheritance from more than one base class.

#include <iostream>
using namespace std;
class B1 {
private :
int m ;
public :
int a;
void setvalue( )
{m = a;}
}; // End of class B1
class B2
{ public :
int k;
}; //End of class B2

class D : public B1,public B2

// Declaration in multiple in heritance
{
public:
int n ;
};
// End of class D
int main()
{ D C; // C is declared an object of class D
C.k = 5; // C accesses directly the public member of B2
C.a = 4 ; // C accesses directly the public member of B1
C.n = 3 ;
int product= C.a* C.n*C.k;

cout<<“Product of data members of B1, B2 and D = ”<< product<< endl;

return 0;
}

The expected output is as follows.

Product of data members of B1, B2 and D = 60

14.7 MULTILEVEL INHERITANCE

The multilevel inheritance is illustrated in Fig.14.4c. The figure shows that class A is the base
class for class B and class B is the base class for class C. The rules of inheritance as discussed
 Inheritance ❖ 353❖

above apply to this as well. This is illustrated in the following program in which class B is the
base class to derived class D1 which in turn is the base class for class D2. All the three classes
contain a function with same name Display. In the main () these functions are called. In order
to remove the ambiguity as to Display function of which class is called, the scope resolution
operator is used for the Display() functions of class B and class D1. For Display function of
class D2 the scope resolution operator is not necessary because the function call of object of
D2 would go to the function of class D2.

PROGRAM 14.10 – Illustrates multilevel inheritance.

#include <iostream>
using namespace std;
class B
{ public :
void Display () { cout << “It is class B Display” <<endl;}
};
class D1 : public B
{
public :
void Display () { cout << “It is class D1 Display” << endl;}
};
class D2 : public D1
{ public :
void Display () { cout << “It is class D2 Display”<< endl;}
};
int main ()
{
D2 d2 ; // d2 is an object of class D2.
d2.B::Display();
d2.D1::Display() ;
d2.Display();
return 0 ;
}

The expected output is given below. The above code also shows that objects of derived class are
also objects of base class.
It is class B Display
It is D1 Display
It is D2 Display
The following example is another instance of multilevel inheritance. In this, instead of
constructor function another public function is used to initialize the objects of different classes.
The constructors and destructors in inheritance are discussed in Section 14.8 of this chapter.
❖ 354 ❖ Programming with C++

PROGRAM 14.11 – Illustrates multilevel inheritance.

#include <iostream>
using namespace std;
class B
{ protected :
int bx ;
public :
void Bset( int m) { bx = m ;}
void Display (){cout<< “Display of class B ”<<endl;}
} ; // End of class B
class D1 : public B
{
protected :
int D1x ;
public :
void D1set(int n){ D1x = n;}
void Display (){cout << “Display of class D1”<<endl;}
} ; // End of class D1
class D2 : public D1
{
protected :
int D2x ;
public :
void D2set( int p ) { D2x = p ;}
void Display (){ cout << “Display of class D2”<<endl;}
}; // End of class D2
class D3 : public D2
{
private :
int D3x;
public :
void D3set ( int q) { D3x = q;}
int Product ( ) { return D3x * D2::D2x * D1::D1x * B::bx ;}
void Display (){ cout << “Display of class D3 ”<<endl;}
};
// End of class D3
int main ()
{
D3 d3;
 Inheritance ❖ 355❖
d3.D3set (10);
d3.D2::D2set(4);
d3.D1::D1set(3);
d3.B::Bset(2);

cout << “Product = ”<< d3.Product()<<endl;

d3.Display();
d3.D2::Display();
d3.D1::Display();
d3.B::Display();
return 0 ;
}

The expected output of the program is as below.

Product = 240
Display of class D3
Display of class D2
Display of class D1
Display of class B

14.8 CONSTRUCTORS AND DESTRUCTORS IN INHERITANCE

In the illustrative examples mentioned above we have used public functions other than
constructors to initialize the class objects. If base classes do not have constructor functions or
their constructors do not take any argument then the derived class need not have a constructor
function. However, if any of the base class has constructor function that takes an argument
then derived class has to have its constructor function because it is the responsibility of derived
class constructor to initialize the base class or supply arguments to base class constructor.
Therefore, the derive class constructor carries data for its own initialization as well as for base
class constructors. An illustration of derived class constructor is given below. Let the classes B1
and B2 be the two base classes and let class D be the derived class. The class B1 has a private
data member int x and B2 has a private data member int y. The constructors of the base
classes are as below.
B1(int n) { x = n ;} // constructor of B1
B2(int m ) { y = m ;} // Constructor of B2
The constructor of derived class D in which int z is a private data member, would be coded
as,
D ( int i, int j ,int k ) : B1(i), B2(j) {z = k;}
See carefully the above expression. The constructor is declared for class D , so the first word
is the name of class, i.e. D. This is followed by the names of parameters preceded by their types
individually in parentheses. Then there is a colon (:) which is followed by name of the first base
class with the parameter (i) in parentheses. The integer i is for the constructor of B1. Similarly
this is followed by comma and then by name of second class B2 and the j in parentheses for
constructor of B2. The int k is meant for class D and is assigned to z in braces {}.
❖ 356 ❖ Programming with C++

In the constructor of class D, the int i and int j are for the constructors of base
classes and int k is for the private member of class D. The order of execution of constructors
and destructors of base and derived classes are listed in Table 14.2. It is obvious that the
constructors of base classes should be called first in the order in which they are declared in
declaration of derived class. Thus if the declaration is as below.
class D : public B1, public B2
the constructor of B1 will be called first followed by that of B2 and in last the constructor of
class D. For destructors this order is reversed. First the destructor of derived class D is called then
that of class B2 and then destructor of class B1. The following program illustrates the above
discussion.

PROGRAM 14.12 – Illustrates definition of constructor in derived class.

#include <iostream>
using namespace std;
class B1{
private :
int x;
public :
B1() { x =1;} // default constructor of B1
B1(int a ) { x = a;} // parametric constructor of B1
int getx(){return x ;} // another public function of B1
~ B1 () { } // destructor for B1
}; // End of class B1

class B2 {
int y ; // private by default

public :
B2(){y=2;}
B2(int b ) { y = b;} // Parametric constructor of B2
~ B2(){ } // Destructor for B2
int gety(){ return y;}
} ; // End of class B2

class D : public B1 , public B2

{
private :
int z;
public :
D ( int i, int j ,int k ) : B1(i), B2(j) { z = k;}
// Constructor for D
~ D () { } // Destructor of D
int Product () {return getx()* gety()*z;} // A function of D
}; // End of class D

int main()
 Inheritance ❖ 357❖
{ D d (4, 5 , 6); //d is an object of class D
cout<< “Product = “ <<d.Product()<<endl;
return 0; }

The expected output is given below.

Product = 120
The order in which the constructors and destructors are executed is illustrated in the
following program.

PROGRAM 14.13 –Illustrates the order of execution of constructors and destructors in

inheritance.
#include <iostream>
using namespace std;

class B1 { // start of class B1

private :
int x;

public :
B1 ( ) {x = 4;
cout << “Constructor of B1 called\n”;} // constructor

~ B1(){cout<<“Destructor of B1 called\n”;} //Destructor of B1

}; // End of class B1

class B2 // start of class B2

{
private:
int y ;
public :

B2 ( ) { y = 5;
cout << “Constructor of B2 called\n”;} //default constructor
~ B2(){cout<<“Destructor of B2 called\n”;} //Destructor of B2} ;
// End of class B2

class D : public B1, B2 // class D derived from B1 and B2 { private :

int z;
public:
D ( int c ) { z =c ; cout<< “Constructor of D called\n” ;}
~ D () { cout<< “Destructor of D called\n”;}
} ; // End of class D

int main()
{
❖ 358 ❖ Programming with C++
D d (10); // d is declared an object of class D
return 0;
}

The expected output is as under.

Constructor of B1 called
Constructor of B2 called
Constructor of D called
Destructor of D called
Destructor of B2 called
Destructor of B1 called

In case there is only one base class, the constructor of base class is called before the constructor
of derived class. If there are several base classes but no virtual base class, the constructor functions
of base classes are called in the order in which the base classes are declared in the declaration of
derived class but before the constructor of derived class.
If one of the base class is a virtual class, its constructor is the first to be called followed by
constructors of other base classes in the order in which they are declared. For example let the
derived class be declared as
class D : public B1, public B2 , public virtual B3
The constructors would be called in the following order.
B3( ) Constructor of virtual class B3
B1( ) Constructor of base class B1
B2( ) Constructor of base class B2
D( ) Constructor of derived class D
The Table 14.2 also gives the order of calling of constructor and destructor functions for a
number of cases.
Table 14.2 – Order of execution of constructors and destructors in inheritance
Category of inheritance Priority in execution of constructor/ destructor in
base and derived classes
class D : public B B( ) Constructor of base class B
D( ) Constructor of derived class D
~ D( ) Destructor of derived class D
~B( ) Destructor of base class B
class D: public B1, public B2 B1( ) Constructor of base class B1
B2( ) Constructor of base class B2
D() Constructor of derived class D
~D( ) Destructor of derived class D
~B2 ( ) Destructor of base class B2
~B1 ( ) Destructor of base class B1
Contd…
 Inheritance ❖ 359❖

class D: public B1, public virtual B2 B2( ) Constructor of virtual class B2

B1( ) Constructor of base class B1
D( ) Constructor of derived class D
~D( ) Destructor of derived class D
~B1 ( ) Destructor of base class B1
~B2 ( ) Destructor of base class B2

COPY CONSTRUCTOR AND INHERITANCE

The following program is an example of copy constructor and inheritance.

PROGRAM 14.14 – Illustrates copy constructor in inheritance.

#include <iostream>
using namespace std;
class B
{ protected:
int x;
public:
B () { x = 4;}
B(int a) {x = a;
cout<< “Base constructor called”<<endl;}
B( const B &b) { x = b.x;
cout<< “copy constructor of B called.”<<endl;}
};
class D : public B
{
int y; // private by default
public:
D () { y = 5;}
D ( int k ){ y = k;
cout<< “Derived class constructor called.”<<endl;}
int Display () { return y ;}
};
int main()
{
D d1 (10); // d1 is defined an object of class D

D d2 = d1; // d2 is defined as a copy of d1.

cout << “d2 = “<<d2.Display() <<endl;

return 0 ;
}
❖ 360 ❖ Programming with C++

The expected output is given below.

Derived class constructor called.
copy constructor of B called.
d2 = 10

The following program is yet another example of declaration of derived class constructor.
In this example the program class Cubicle calculates the surface area of prismatic bodies having
three integer dimensions. This class is a base class to the class Cube which uses the surface area
for finding the cost of painting the surface. Examine the constructor by name Cube in the
following program.

PROGRAM 14.15 – Illustrates constructors of base and derived classes.

#include <iostream>
using namespace std;

class Cubicle {

private:
int x, y, z;
public:

Cubicle (){}
Cubicle (int a, int b, int c ) // Constructor
{x = a;
y=b;
z=c;

cout<<“constructor of Cubicle called”<<endl;}

int surface_area() { return 2*(x*y +y*z +z*x);}
}; // end of class Cubicle

class Cube : public Cubicle

{
public:
int Rate;
Cubicle C;
// below is the constructor function of class Cube
Cube(int L, int W, int H, int A): Cubicle (L, W, H) { Rate = A;}

int cost(int Rate, Cubicle C ){return Rate * C.surface_area() ;}

}; // end of class Cube

int main()
{
int x =2;
Cubicle C1 (3, 3, 3);
Cubicle C2 (4, 5, 6);
 Inheritance ❖ 361❖
Cube cube1(3,3,3, x);
Cube cube2(4,5,6,3);
cout<<“Surface area of cube1 = ”<<cube1.surface_area()<<“\n”;
cout << “Cost for cube1 = ”<< cube1.cost(x, C1 )<<“\n”;
cout<<“Surface area of cube2 = ”<<cube2.surface_area()<<“\n”;
cout<<“Cost for cube2 = ” << cube2.cost(3,C2)<<“\n”;
return 0 ;
}

The expected output is given below.

Constructor of Cubicle called
Constructor Cubicle called
Constructor of Cubicle called
Constructor of Cubicle called
Surface area of cube1 = 54
Cost for cube1 = 108
Surface area of cube2 = 148
Cost for cube2 = 444

14.9 CONTAINMENT AND INHERITANCE

For many real life problems we have classes that contain objects of other classes as their members
besides having their own function members and data members. This is a form of nesting called
containership or container class. The other form of nesting is that a class is declared inside another
class. The container class will have characteristics of contained classes through their objects. C++
also supports such a relationship. In a way, this provides an alternate method to inheritance. Thus
if a class D has objects of class B and class C as its members we say D has a relationship with C and
B. Here we shall discuss how to define the constructor of the container class. Let the three classes
B , C and D be defined as below. Class D contains objects of class B and class C.
class C{
int x; // private by default
public:
C ( int a ) { x=a ;} // Constructor of class C
———
}; // End of class C
class B
int y; // private by default
Public:
B ( int b ) { y = b;} // constructor of class B
——————
}; // end of class B
class D{
int z;
public:
C object_C;
B object_B;
// Constructor of class D is given below.
❖ 362 ❖ Programming with C++
D(int i, int, j, int k) : object_C (i), object_B(j){z = k;}
——
}; // end of class D

In the above definition, the constructor for D has three integers in parentheses followed by
colon (:) for the three constructors. The constructors of class B and class C are called by their
objects, i.e. object_B and object_C which follow the colon and contain the arguments in
parentheses for their constructors, this is followed by constructor body for class D, i.e.
{ z = k ;}. The following program illustrates this.

PROGRAM 14.16 – Illustrates constructor of a class containing objects of other classes.

#include <iostream>
using namespace std ;

class C
{
private :
int x;
public :
C(int a) {x = a;
cout << “Constructor of C called.\n”;} // Constructor
~ C () { cout<<“Destructor of C called\n”;} //destructor
int getx( ) { return x;}
}; // End of class C

class B {
private:
int y ;

public :

B (int b) { y = b;
cout << “Constructor of B called\n”;} // Constructor
~ B () { cout<<“Destructor of B called\n”;} // Destructor
int gety(){return y;}
}; // End of class B

class D
{
private :
int z;
public:
C objC ; // objC is declared an object of Class C
B objB ; // objB is declared an object of class B
 Inheritance ❖ 363❖
D ( int i, int j, int k ) : objB (i) , objC(j) { z =k ; cout<< “Constructor of
D called\n” ;}
// Constructor of class D
~ D () { cout<< “ Destructor of D called\n”;} //Destructor
int Product (){return objC.getx()* objB.gety()* z;}
}; // End of class D
int main()
{
D objD(5, 20, 4 );
cout << “Product = ” << objD.Product () <<endl;
return 0;
}

The expected output is as under.

Constructor of C called.
Constructor of B called
Constructor of D called
Product = 400
Destructor of D called
Destructor of B called
Destructor of C called

A derived class may contain objects of other classes, thus presenting a case of inheritance as
well as containment. The derived class constructor feeds arguments to the constructors of base
class as well to those classes whose objects are contained by derived class. For the contained class
the name of its object is used while for base class the name of class is used to feed arguments to
the corresponding constructor. The following program illustrates this.

PROGRAM 14.17 –Illustrates constructor of derived class having a base class and
containing object of another class.

#include <iostream>
using namespace std;
class C
{
private :
int x;
public :
C (int a) {x = a;
cout << “Constructor of C called.\n”;} //constructor of class C
~ C () { cout<<“ Destructor of C called\n”;}
int getx( ) { return x;}
}; // end of class C
❖ 364 ❖ Programming with C++
class B {
private:
int y ;
public :
B (int b) { y = b;

cout << “Constructor of B called\n”;} // constructor of class B

~ B () { cout<<“ Destructor of B called\n”;}

int gety(){return y;}

}; // End of class B

class D : public B // class D derived from class B

{
private :
int z;
public:
Cc; // Object c of class C in class D
// Below is the constructor of class D.
D ( int m, int n, int k ) : B(m) , c(n) { z =k ;
cout<< “Constructor of D called\n” ;}

//constructor of B called by class name B.

// constructor of C called through name c of its object

~ D () { cout<< “ Destructor of D called\n”;}

int Product () { return c.getx()* gety()* z;}
} ; // End of class D

int main()
{
D d (10, 20,5); // object of class D
cout << “Product = ” << d.Product () <<endl;
return 0;
}

The expected output is shown below. Examine the priority in which the constructors and
destructors are called.
Constructor of B called
Constructor of C called.
Constructor of D called
Product = 1000
Destructor of D called
Destructor of C called
Destructor of B called
 Inheritance ❖ 365❖
Program 14.18, given below, is yet another example of declaration of constructor function
in a multilevel inheritance. The method of constructing the constructor function is similar to
the class having both the base class as well as objects of other classes.

PROGRAM 14.18 – Illustrates constructor function for a derived class in a multilevel

inheritance.

#include <iostream>
using namespace std;
class B {
protected :
int bx ;
public :
B (){}
B( int m) { bx = m ;}
int getbx(){ return bx;}
} ; // End of class B
class D1
{
protected :
int D1x ;
public :
D1(){}
D1( int m ) { D1x = m;}
int getD1x () { return D1x;}
};
class D2 :public D1
{
protected :
int D2x ;
public :
int getD2x(){return D2x;}
D2 (){}
D2( int e ) { D2x = e ;}
};
class D3 : public D2
{
private :
int D3x;
public :
B b ; // b is an object of class B
❖ 366 ❖ Programming with C++
D1 d1; // d1 is an object of class D
// The constructor of D is defined below.
D3 (int i, int j, int k, int s ) : b(i), d1(j), D2(k) { D3x = s;}
int Product () { return D3x * D2x * d1.getD1x() * b.getbx() ;}
void Display (){ cout << “Display of class D3 ”<<endl;}
};
int main ()
{
D3 d3 ( 10, 5, 4, 2);
cout << “ Product = ”<< d3.Product( )<<endl;
return 0 ;
}

The expected output is as below.

Product = 400

14.10 OVERLOADED OPERATOR FUNCTIONS AND INHERITANCE

The overloaded function of base class is inherited by derived class. This is illustrated by the
following program in which the overloaded operator function += is defined in the base class.
The function is invoked when it is applied to objects of derived class.

PROGRAM 14.19 – Illustrates inheritance of overloaded function +=.

#include<iostream>
using namespace std;
class Vector
{
protected:
int x, y, z ; // three components of vector
public :
Vector (){ x=1,y=2,z=3;}
Vector (int a, int b,int c) { x =a ; y = b; z =c;
cout<< “constructor of class Vector called.”<<endl;}
Vector operator += (constVector & P)
{
x = x + P.x ;
y = y + P.y ;
z = z + P.z ;
cout << “overloaded function of Vector called.”<<endl;
return *this;}
void Display() {
 Inheritance ❖ 367❖
cout << “x component = ”<< x << endl;
cout << “y component = ”<< y << endl;
cout << “z component of = ”<< z << endl;}
} ; // End of class Vector
class D : public Vector
{
public :
// constructor function
D ( int a, int b, int c ): Vector (a,b,c) {
cout<<“constructor of D called.”<<endl;}
};
int main()
{
D V1 ( 12,4,6), V2( 3, 5, 7) ;
V1 += V2;
V1.Display( ) ;
return 0;
}

The expected output given below shows that overloaded functions are inherited.
constructor of class Vector called.
constructor of D called.
constructor of class Vector called.
constructor of D called.
overloaded function of Vector called.
constructor of class Vector called.
x component = 15
y component = 9
z component of = 13

FRIEND FUNCTIONS AND INHERITANCE

The friend functions are not inherited by the derived class.

EXERCISES
1. What is inheritance? How is it useful in object oriented programming?
2. What are the different types of inheritances?
3. What is the difference between single inheritance and multiple inheritance. Explain with diagram
and class declarations.
4. With the help of a diagram show the multi-path inheritance. Also give an example of derived
class declaration.
❖ 368 ❖ Programming with C++
5. In case of single inheritance distinguish between public and private inheritances?
6. In a single inheritance how the public inheritance and protected inheritances are different?
7. What is the difference between multiple inheritance and multilevel inheritance? Explain with
diagrams and class declarations.
8. How do you access the public and protected members of base class in a public inheritance?
9. In Fig. 14.2 how would you declare the class Hexagonal_headed_bolts as a derived class?
10. How can an object of derived class access the private function of base class?
11. A derived class D is derived from two base classes B1 and B2 which have one private data
member each and constructor functions. The derived class D also has a private member. How
would you define the constructor function for class D.
12. What will be the order of execution of constructors in following class declarations?
(i) class D : public class B1, public B2
(ii) class D : public virtual class V, private class B1, public class B2
(iii) class D : public B1, public B2, public virtual V.
13. Give an example of constructor of derived class for a single public inheritance in which the base
class has one integer private data member and derived class has one private data member
of type double.
14. For a multilevel inheritance, construct the constructor function of class D2 which is derived from
class D1 which in turn is derived from class D which in turn is derived from class B. All the
classes except D2 have one protected data member each while D2 has a one private data
member.
15. Write a program to declare a derived class by name Grades from the base class Student. The
base class has roll numbers and names while the grade class has roll numbers and grades.
The derived class displays the marks list giving students’ names, roll numbers and grades.
16. What is containment? How would you define the constructor function of a class which contains
objects of two other classes?
17. A class D is derived from a class B and contains the object of another class C. If all the
classes have constructor functions how would you define the constructor function of class D?
18. Do you need a constructor function if the base and derive class have only public function?
19. What is the order of calling of destructor function in a multilevel inheritance?
20. Do you need a constructor function in the derived class if the constructor functions of base
classes do not take arguments.
21. Give an illustration of a constructor function of a class which is derived with single public
inheritance and has objects of other two classes as its members.

❍❍❍
CHAPTER

15.1 INTRODUCTION

Polymorphism is a combination of two Greek words, i.e. poly and morphism, which means
the ability to acquire different forms or shapes according to the environment. We have already
got acquaintance with two forms of polymorphism which are function overloading and operator
overloading. In operator overloading the same operator carries out different forms of operations
according to the type of data presented to it. In function overloading several functions have the
same name but they differ in their number of arguments or types of arguments. So the same
function name is called to carry out different actions. Templates (see Chapter 16) is still another
form of polymorphism. All these cases are examples of static binding or early binding because
the choice of the appropriate function gets decided at the compile time itself that is before the
execution of program starts.
The polymorphism in which the choice of function is done during the execution of the
program is called run time binding or late binding or dynamic binding. This is achieved through
virtual functions and pointer to the base class. All these are illustrated in Fig. 15.1 below.

Static binding Function overloading

or
compile time
Polymorphism binding Operator overloading

Dynamic binding or Virtual functions and

run time binding base class pointers

Fig. 15.1: Polymorphism

15.2 VIRTUAL FUNCTIONS

The run time polymorphism is realised through virtual functions. But the application of virtual
functions is through the base class pointer. The reason for using base class pointer is illustrated
in the Fig. 15.2(a, b). A base class pointer can be made to point to objects of base class as well
as to objects of derived classes. But a derived class pointer can point to objects of the derived
class only. It cannot point to objects of base class because a base class object is not an object of
derived class. If another class is derived from a derived class, i.e. class Derived2 from class
❖ 370 ❖ Programming with C++

Derived1 as shown in the figure below, then the class Derived1 is in fact a base class to class
Derived2. In that case the pointer to class Derived1 can point to objects of class Derived1 and
objects of class Derived2. But pointer to class Derived2 cannot point to objects of class Derived1.
However, the class Base pointer (see Figure15.2b) can point to objects of Base class as well as
to objects of class Derived1 and to objects of class Derived2. Thus with pointer of a base class
we can access the objects of the whole hierarchy of objects on the down stream.
class Base pointer

Base class object

class Base Derived1 class object

class Derived1 pointer

class Derived1

Derived2 class object

class Derived2
class Derived2 pointer
(a) (b)

Fig. 15.2: Class pointers and their reach

In order to clearly bring out the characteristics of virtual functions, the following four
programs have been designed. In Program 15.1 a base class with name B and a derived class with
name D have been defined (class D : public B). Both the classes have got a public function by
name Display( ). In the base class it is defined to display “Are you going to learn C++?” and in
derived class it is defined to display “I am also learning C++.”. Pointer bptr is pointer to class
B. It is initialized by &(B)b where b is an object of class B. When this pointer is used to call
the function Display () , naturally it displays “Are you going to learn C++?”, i.e. base class
definition.
Now the pointer is assigned a new address &d where d is an object of the derived class D.
The object d is also an object of base class B because of inheritance. The function Display() is
again called by the pointer. It again displays the same “Are you going to learn C++?” and not the
definition given in class D. See the following program for illustration.

PROGRAM 15.1 – Illustrates that if a function is not declared virtual the base class pointer
will point to base class function definition even though it is assigned the
address of derived class object.

#include <iostream>
using namespace std;
class B
{
 Virtual Functions and Polymorphism ❖ 371❖
public:
void Display ()
{cout<< “Are you going to learn C++?”<<endl;}
}; // End of class B
class D : public B
{
public :
void Display( )
{cout<< “I am also learning C++.”<<endl;}
}; // End of class D
int main()
{ B b; //b is an object of base class B
D d; // d is an object of derived class D
B *bptr; // pointer to class B
bptr = &(B)b; //pointer initialization of pointer to B
bptr -> Display(); // Function called by pointer
bptr = & d; //the pointer is assigned address of d.
bptr -> Display (); //the pointer is used to call the function
return 0;
}

The expected output is given below.

Are you going to learn C++?
Are you going to learn C++?
From the output it is clear that even though the pointer bptr of class B has been assigned
the address of object d of derived class D, the function displayed through the pointer is the
function of base class.
Now if the pointer is defined for class D and initialized by value & (D) d and if we try to
assign to it the address of base class object, i.e. &b, the compiler will show an error because b is
not an object of class D. The objects of derived classes are also objects of base class but
objects of base class are not the objects of derived class. In fact, a base class does not know
the existence of any class derived from it. This is illustrated in the following program.

PROGRAM 15.2 – Illustrates that pointer of derived class cannot point to an object of base
class.

#include <iostream>
using namespace std;
class B
{
public:
❖ 372 ❖ Programming with C++
void Display ()
{ cout<< “Are you going to learn C++?”<<endl;}
}; // End of class B

class D : public B
{
public :
void Display( )
{cout<< “I am also learning C++.”<<endl;}
}; // End of class D

int main()

{ B b; // b is declared an object of class B

D d; // d is declared an object of class D
D *dptr = &(D)d; // pointer to class D
dptr -> Display();

dptr = & b; // dptr is assigned address of base class

// this would generate an error message.
dptr -> Display ();
return 0;
}

The expected output is the following error message.

cannot convert from class B* to class D*
The above result is due to the fact that objects of base class are not objects of derived class.
Therefore, a derived class pointer cannot call a function of base class.
Program 15.3 given below shows that the appropriate form of function Display () may be
obtained by defining separate pointers for class B and class D. Though we get the relevant
functions but that is not what is desired. We wish to point to object of the derived class by pointer
of base class so as to get the form of function Display() defined in derived class D.

PROGRAM 15.3 – Illustrates the actions of separate pointers of base class and of derived
class.
#include <iostream>
using namespace std;
class B
{
public:
void Display ()
{ cout<< “Are you going to learn C++?”<<endl;}
} ; // End of class B
 Virtual Functions and Polymorphism ❖ 373❖
class D : public B
{
public :
void Display( )
{cout<< “I am also learning C++.”<<endl;}
} ; // End of class D

int main()
{Bb; //b is an object of base class B
Dd; // d is an object of derived class D
B *bptr = & (B) b ; //pointer to B
D *dptr = &(D) d ; // pointer to class D

bptr -> Display ();

dptr -> Display();

return 0;
}

The expected output is as under.

Are you going to learn C++?
I am also learning C++.
The above output is obtained by defining two different pointers, i.e. one for base class and
the other for derived class. It is natural to get the respective form of function Display (), but it
not our aim. Aim is to get the corresponding derived class functions by base class pointers. This
can be achieved only by declaring the function Display() a virtual function in the base class. The
word virtual is a keyword in C++. In the derived class the word virtual may or may not be used
but the prototype of the virtual function in the base class should be identical with the prototypes
of the function in the derived classes. In the following program the function void Display()
has been declared virtual. The definition of function Display () is modified to following type.
virtual void Display () { statement; }
Now the same function is called in derived class through base class pointer. The definition
given in the derived class overrides the base class definition, and in each derived class its own
definition is carried out.
This is illustrated in Program 15.4 given below. In the derived class the function should have
same name and same parameter types as that of base class virtual function. If the type or number
of parameters are changed it would be taken as overloaded function and it would lose the property
of virtual function. One may or may not use the keyword virtual in the derived class. But the
corresponding function in derived class should have the same name and same parameter types,
i.e. the prototypes of the function in the base class and in the derived classes should be identical.
❖ 374 ❖ Programming with C++

PROGRAM 15.4 – Illustrates application of virtual function.

#include <iostream>
using namespace std;
class B
{
public:
virtual void Display ()
{ cout<< “Are you going to learn C++?”<<endl;}
}; // End of class B
class D : public B
{ public :
void Display( )
{cout<< “I am also learning C++.”<<endl;}
}; // End of class D
int main()
{
B b;
D d;
B *bptr = &(B) b ; // Base class pointer
bptr -> Display(); // statement for base class function
bptr = & d; // base class pointer assigned address of d
bptr -> Display (); // calling function Display ()
return 0;
}

The expected output is given below. Note that the function Display() has been declared
virtual. The base class pointer has been assigned the address of object of derived class and then
the pointer is used to call the function Display(). In this case the derived class definition of the
function Display () overrides the base class definition of the function and the result is shown in
the output given below.
Are you going to learn C++?
I am also learning C++.

CHARACTERISTICS OF VIRTUAL FUNCTIONS

(1) A virtual function should be defined in the class in which it first appears (base class). It
is defined as a public member function with the keyword virtual preceding the definition.
Exception to this is the pure virtual function, in which case, there is no definition of the
function in the base class. See Section 15.4 below.
(2) A virtual function cannot be a static member.
(3) A virtual function may be declared an inline function.
(4) A virtual function may be declared a friend of another class.
 Virtual Functions and Polymorphism ❖ 375❖

(5) Apart from the same name, the virtual function prototypes in the base and derived classes
should match exactly in terms of type, number of arguments and respective types of
arguments. Any deviation from this will make the compiler to take it as an overloaded
function and it will lose the status of virtual function.
(6) A constructor function cannot be a virtual function.
(7) A destructor function may be a virtual function.
(8) The virtual function definition in base class is overridden by the respective definitions
in the derived classes. If in the derived class the function is not redefined, then the base
class definition is invoked.
(9) The virtual functions are inherited to any depth of hierarchy.
(10) A class containing one or more virtual functions is often called polymorphic class.
The following program illustrates the application of a virtual function.

PROGRAM 15.5 – Illustrates polymorphism.

#include <iostream>
using namespace std;
class B
{public :
virtual void Display(){cout<<“Display of class B called”<<endl;}
}; // End of class B
class D1 : public B
{ public :
void Display(){cout << “Display of class D1 is called”<<endl;}
}; //End of class D1
class D2 : public B
{
public :
void Display (){ cout << “Display of class D2 is called”<<endl;}
} ; // End of class D2
int main ()
{
B b; // b is an object of class B
B *bptr = &b; // base class pointer
D1 d1; // d1 and d2 are objects of classes D1
D2 d2; // and D2 respectively.
bptr ->Display() ;
bptr = &d1; // base class pointer assigned address
// of derived class object
bptr -> Display ();
❖ 376 ❖ Programming with C++

bptr = &d2;
bptr ->Display() ;
return 0 ;
}

The expected output is given below.

Display of class B called
Display of class D1 is called
Display of class D2 is called

15.3 ARRAYS OF BASE CLASS POINTERS

When there are several derived classes which redefine and use the virtual function of a base class,
it is convenient to declare an array of pointers to the base class. The elements of array are the
addresses of respective class objects. Let the classes such as D1, D2, D3, etc., be derived from a
base class B. Let b, d1, d2, d3, etc. be the objects of classes B, D1, D2, D3, etc. For such a
situation we may declare and initialize an array of base class pointers as below.
B *bptr[] = {&b, &d1, &d2, &d3};
where &b is the address of base class object and &d1, &d2 and &d3 are the addresses of the
objects of derived classes D1, D2 and D3 respectively. The application is illustrated in the
following program.

PROGRAM 15.6 – Illustrates an array of base class pointers to base and derived class objects.

#include <iostream>
using namespace std;
class B
{
public :
virtual void Display(){cout<< “Display of class B called”<<endl;}
}; // End of class B
class D1 : public B
{ public :
void Display (){ cout << “Display of class D1 is called”<<endl;}
}; //End of class D1
class D2 : public B
{
public :
void Display (){ cout << “Display of class D2 is called”<<endl;}
}; //End of class D2
class D3 : public B
 Virtual Functions and Polymorphism ❖ 377❖
{ public :
void Display (){ cout << “Display of class D3 is called”<<endl;}
}; //End of class D3
int main ()
{B b;
D1 d1;
D2 d2;
D3 d3;
B *bptr[] = { &b , &d1, &d2, &d3}; // Array of pointers of B
for ( int i = 0; i<4; i++)
bptr[i] ->Display() ;
return 0 ;
}

The expected output is given below.

Display of class B called
Display of class D1 is called
Display of class D2 is called
Display of class D3 is called

In the above program the elements of an array of pointers to the base class are assigned the
values which are the addresses of the objects of base class and derived classes. In the output we
see that the function definitions of the respective classes are called.

15.4 PURE VIRTUAL FUNCTIONS AND ABSTRACT CLASS

A pure virtual function is not defined in the base class in which it first appears but is defined in
the derived classes. In the base class it has no function body and its declaration is followed by =
0. The declaration of a pure virtual function in base class is illustrated below.
virtual type Identifier (type parameters, —,—) = 0 ;
Here virtual is a keyword of C++, type is the type of return data and identifier is the name
of function. There is no definition of the function in the base class. The prototype of the function
is equated to zero. But the virtual function is defined in derived classes in their own way. A class
containing a pure virtual function is called abstract class. An abstract class cannot have objects
of its own. The following are the consequences of declaration of pure virtual function.
(i) A pure virtual function cannot be used for any action in the base class. In fact, the base
class with a pure virtual function cannot have objects and hence is not initialized.
(ii) The class in which one or more pure virtual functions are declared is known as abstract
class or pure abstract class.
(iii) An abstract class cannot have any objects of its own.
(iv) The derived classes have their own definitions of the pure virtual function in the respective
classes.
(v) If a derived class does not define the pure virtual function declared in its base class, it also
becomes an abstract class.
❖ 378 ❖ Programming with C++

The following program gives an illustration of abstract class and the compiler gives an error
message if objects to the abstract class are declared.

PROGRAM 15.7 – Illustration of pure virtual function.

#include <iostream>
using namespace std;
class B
{ public :
virtual void Display () = 0;
}; // End of class B
class D1 : public B // Derived class
{ public :
void Display (){ cout << “Display of class D1 is called”<<endl;}
};

int main()

{ B b ; // object b of B declared but it will result in

// an error message.
B *bptr = &b;
bptr -> Display();
return 0 ;
}

The output is an error message that an abstract class cannot have an object. It shows that an
attempt to declare objects of abstract class would result in compiler error given below.
cannot initiate abstract class.
The following program illustrates an application of pure virtual function. A class dealing with
regular shapes has a pure virtual function Area. The function Area is defined in class Circle and
in class Square which are the derived classes of class Regularshapes.

PROGRAM 15.8 – Illustrates another application of pure virtual function.

#include <iostream>
using namespace std;
class Regularshapes {
int side; // private by default
public:
void set_side( int a )
{ side = a;}
int getside () { return side;}
virtual double Area (void) =0;
}; // end of class Regularshapes
 Virtual Functions and Polymorphism ❖ 379❖
class Circle : public Regularshapes
{
public:
double Area (void)
{return (3.14159*getside()*getside()/4) ;}
}; // end of class Circle

class Square : public Regularshapes

{
public:
double Area (void)
{return (getside()*getside()) ;}
}; // end of class square

int main ()
{ double A1, A2;
Square S1;
Circle C1;
Regularshapes* RS1 = & S1;
Regularshapes* RS2 = & C1;

(*RS1).set_side ( 10);
RS2 -> set_side (10);
A1 = (*RS1). Area(); // (*RS1).Area() and RS1 ->Area() are
// equivalent
A2 = RS2 -> Area ();
cout << “A1 = ” << A1 << “\t A2 = ”<< A2<<endl;
return 0;
}

The expected output is given below.

A1 = 100 A2 = 78.5397

15.5 VIRTUAL DESTRUCTORS

Destructor function may be declared virtual while constructor function cannot be declared
virtual. It is natural to ask, what is the benefit of declaring destructor function a virtual function?
In case of dynamically created objects of a derived class through base class pointer, if the base class
pointer is deleted only the base class destructor is invoked. This is illustrated in the following
program.
However, if the base class destructor function is declared virtual, on deletion of base class
pointer the destructor functions of derived class as well as related classes (classes contained in
derived class) would also get invoked. The following two programs illustrate it.
❖ 380 ❖ Programming with C++

PROGRAM 15.9 – Illustrates behaviour of destructor functions when base class destructor
is not declared virtual.

#include <iostream>
using namespace std;
class B
{
protected :
int bx ;
public :
B () {} // empty constructor of base class
B (int m) { bx = m ; // constructor of base class
cout<< “Constructor of B called” <<endl;}
virtual int Product ( ){return 2*bx;}
// The destructor function not declared virtual
~ B () { cout << “Destructor of class B called”<<endl;}
} ; // End of class B
class D1
{
protected :
int D1x ;
public :
D1 () {} // empty / default constructor of D1
D1(int n){ D1x = n; // constructor of class D1
cout<< “Constructor of D1 called. ”<<endl;}
int getD1x(){ return D1x;}
~ D1 () { cout<<“Destructor of D1 called”<<endl;}
};
class D2
{
protected :
int D2x ;
public :
D2(){}
D2 ( int p ) { D2x = p ;
cout << “Constructor of D2 called “ << endl;}
int getD2x(){ return D2x;}
~ D2 () { cout <<“Destructor of D2 called.”<<endl;}
};
class D3 : public B
{
int D3x; // private by default
 Virtual Functions and Polymorphism ❖ 381❖
public :
D3 () {}
D1 d1; // d1 is declared object of class D1
D2 d2; // d2 is declared object of class D2

D3( int p, int q , int r, int s): B(p), d1(q), d2(r) { D3x = s;
cout << “Constructor of D3 called.”<< endl;}

int Product (){ return D3x * d2.getD2x() * d1. getD1x() * bx ;}

~ D3 () { cout<<“Destructor of D3 called.”<<endl;}
};

int main ()
{
B *Bp ;
Bp = new B(10); // dynamic object created by new

cout << “Product of B = ” <<Bp ->Product ()<< endl;

Bp = new D3( 2,3,4,5);
cout<< “Product = ”<<Bp -> Product() << endl;

delete Bp; // deletion of base class pointer

return 0 ;
}

The expected output is given below.

Constructor of B called
Product of B = 20
Constructor of B called
Constructor of D1 called.
Constructor of D2 called
Constructor of D3 called.
Product = 120
Destructor of class B called

In the output of above program it is clear that destructor of only base class is called. In the
following program the destructor function of base class is declared virtual. With this, on deletion
of base class pointer the destructor functions of derived as well as of related classes (D1 and D2
in the above case) are also invoked. The output of following program illustrates it.
❖ 382 ❖ Programming with C++

PROGRAM 15.10 – Illustrates the effect of declaring base class destructor virtual.

#include <iostream>
using namespace std;
class B
{ protected :
int bx ;

public :
B () {}
B ( int m) { bx = m ;

cout<< “Constructor of B called” <<endl;}

virtual int Product ( ){return 2*bx;}
virtual ~ B (){ cout << “Destructor of class B called”<<endl;}

// Destructor function declared virtual

} ; // End of class B

class D1 // class D1 is not derived from class B but it is a

{ // related class because its object is used in class D3
protected :
int D1x ;

public :
D1 () {}
D1(int n){ D1x = n;
cout<< “Constructor of D1 called. ” <<endl;}
int getD1x(){ return D1x;}

~ D1 () { cout<<“Destructor of D1 called”<<endl;}
};
class D2 // it is also a related class like class D1
{
protected :
int D2x ;
public :
D2(){}

D2 ( int p ) { D2x = p ;

cout << “Constructor of D2 called ” << endl;}

int getD2x(){ return D2x;}

~ D2 () { cout <<“Destructor of D2 called.”<<endl;}

};

class D3 : public B
{
private :
int D3x;
 Virtual Functions and Polymorphism ❖ 383❖
public :
D3 () {} // empty/default constructor
D1 d1; // d1 is object of class D1
D2 d2; // d2 is object of class D2
D3(int p, int q , int r, int s): B(p), d1(q), d2(r) { D3x = s;
cout << “Constructor of D3 called.”<< endl;} // constructor
int Product (){return D3x * d2.getD2x() * d1. getD1x() * bx ;}
~ D3 () { cout<<“Destructor of D3 called.”<<endl;}
};
int main ()
{
B *Bp ;
Bp = new B(10);
cout << “Product of B = ” <<Bp ->Product ()<< endl;
Bp = new D3( 2,3,4,5);
cout<< “Product = ”<<Bp -> Product() << endl;
delete Bp;
return 0 ;
}

The expected output is given below. The destructor functions of base class, derived class and
related classes are called.
Constructor of B called
Product of B = 20
Constructor of B called
Constructor of D1 called.
Constructor of D2 called
Constructor of D3 called.
Product = 120
Destructor of D3 called.
Destructor of D2 called.
Destructor of D1 called
Destructor of class B called

15.6 VIRTUAL BASE CLASS

Figure15.3 below shows a case of inheritance in which two classes D1 and D2 are derived from
a base class B. Another class D3 is derived from these two derived classes. Now an object of class
D3 calls a function of class B. The compiler shows an error that there is ambiguity. Because, both
the classes D1 and D2 have a copy each of class B. Thus D3 has two copies of B. The compiler
cannot decide which of the two copies of B to invoke. The ambiguity can be resolved if the base
class is declared virtual in the definitions of class D1 and class D2. In that case there would be
❖ 384 ❖ Programming with C++

only one copy of the base class in D3 and hence there is no ambiguity. This is taken up in
programs 15.11 and 15.12. In Program 15.11 the ambiguity is illustrated and in Program 15.12
it is resolved.

class Base

class D1 class D2

class D3

Fig 15.3: Multi-path inheritance from same base class

PROGRAM 15.11 – Illustrates ambiguity created in class derived from multi-path derived
classes from same base class

#include <iostream>
using namespace std;
class B
{ public :
void Display () { cout << “ It is B class Display”<< endl;}
};
class D1 : public B
{};

class D2 : public B
{ };
class D3 : public D2, public D1
{ };

int main ()

{ D3 d3;
D3.Display();
return 0 ;
}

The expected output is as below.

‘D3::Display’ is ambiguous
The above ambiguity is resolved by declaring the base class virtual in the declaration of derived
classes.
 Virtual Functions and Polymorphism ❖ 385❖
PROGRAM 15.12 – Illustrates virtual base class.
#include <iostream>
using namespace std;

class B
{
public :
void Display () { cout << “It is B class Display”<< endl;}
};

class D1 : virtual public B

{};

class D2 : virtual public B

{ };

class D3 : public D2, public D1

{ };

int main ()
{
D3 d3;
d3.Display();
return 0 ;
}

The expected output is given below. The ambiguity is resolved by declaring the base class virtual.
It is B class Display

15.7 RUN-TIME TYPE INFORMATION ( RTTI )

As explained in the above sections the pointer to base class can be made to point to objects of
any derived class on the down stream, even to object of classes derived from derived classes.
However, the pointer does not carry any information as to which class object it is pointing to.
In some programs this information is required if we want to apply a special function to objects
of a particular class only. This involves distinguishing the types of objects and finding objects
of the desired class to which the function can be applied.
One solution is to define a member function in each class to supply the type information
and which may be used to select the type of object desired. When the number of classes is large
the coding for selection may become cumbersome with if-else chains or switch expressions. The
virtual functions may also be used to do this job. The following program uses virtual function
Type () to have run time type information.

PROGRAM 15.13 – Obtaining run time type information.

#include <iostream>
using namespace std;
class B
❖ 386 ❖ Programming with C++
{ public:
virtual char Type() { return ‘B’;}
};

class C : public B
{ public:
char Type (){ return ‘C’;}
};

class D : public B
{public:
char Type (){ return ‘D’;}
};

int main()
{ B b;
Cc;
Dd;

B* pb = &b;
cout <<“Type of pb now is : ”<< pb -> Type()<<endl;

pb = & c;
cout<< “Type of pb now is : ”<<pb -> Type()<<endl;

pb = & d;
cout <<“Type of pb now is : ”<<pb -> Type()<<endl;
return 0 ;
}

The expected output is given below. The type information is obtained at run time.
Type of pb now is : B
Type of pb now is : C
Type of pb now is : D

The C++, however, offers direct support to RTTI with two operators, i.e. dynamic_cast ()
and typeid (). The operator typeid ( ) allows its application to polymorphic classes as well to non-
polymorphic classes (see application of typeid in Chapter 2). It determines and returns a reference
to the exact type of the object. The name of the class of the object may also be obtained by
function name(). Figure 15.4 gives a physical illustration of typeid( ) operator.
 Virtual Functions and Polymorphism ❖ 387❖

class Cylinder class Cubic

Operator
typeid ()

Fig. 15.4: Identification of class of objects

The classes relating to typreid () are contained in the header file <typeinfo> and this header
file should be included in the program if its objects are to be used in the program. The syntax
for typeid () is as follows.
typeid (d) ;
where d is an object of a class. The code gives reference to the type of object d. The header file
<typeinfo> also defines two relational operators, i.e. == and != for comparing the types of two
objects say d1 and d2. The syntax is illustrated below.
if(typeid (d1) == typeid (d2))
or
if (typeid(d1) != typeid(d2))
The header file <typeinfo> also defines function name() which gives the name of the type
or name of class to which an object belongs. The syntax for using name() is given below.
typeid(d1).name( );
The function returns the name of the type of object d1. For display, it may be coded as
below.
cout << typeid(d1).name( );
The application of operator typeid () is illustrated in the following program.

PROGRAM 15.14 – Illustration of application of typeid()on non-polymorphic classes.

#include <iostream>
# include <typeinfo> // header file for typeid()
using namespace std;
class B
{ };
class D1 : public B
{ };
class D2 : public D1
{ };
int main ()
{ D2 d2 ;
D1 d1 ;
❖ 388 ❖ Programming with C++
if ( typeid( d1) == typeid (d2))
cout<< “The two are of same type ”<<endl;
else
cout<< “The two are of different type”<<endl;
cout << “Type of d1 is ” <<typeid(d1).name( )<<endl;
cout << “Type of d2 is ” <<typeid(d2).name( )<<endl;
return 0;
}

The expected output is given below. It is self explanatory.

The two are of different type
Type of d1 is class D1
Type of d2 is class D2

15.8 NEW CASTING OPERATORS IN C++

THE DYNAMIC_CAST <> ()

The operator is used to cast types of objects at run time and is applicable only to polymorphic
classes, i.e. the base class should have at least one virtual function. The operator dynamic_cast<>
() is applicable to pointers as well as references. For pointers it is coded as below
dynamic_cast < derived_type*> (ptr_base);
The dynamic_cast converts the base class pointer (ptr_base) to the derived class pointer
(derived_type*) subject to the condition that the object to which the ptr_base is pointing belongs
to the desired derived class or to a class derived from the desired derived class, otherwise it returns
0. So it is a convenient way to select the objects of the desired derived class from objects belonging
to several derived classes. Its physical illustration is shown in the Fig. 15.5 below. The pointers
to several objects from which we wish to select objects of a particular class, are all cast by
dynamic_cast. The pointers which point to objects of desired class will be converted while others
would return 0, so they would be out of list. The following figure gives a physical illustration.

Entry for OP class objects only

OP type

Fig. 15.5: An illustration of dynamic_cast <> ()

 Virtual Functions and Polymorphism ❖ 389❖

An illustration of code for dynamic_cast <> () is given below.

Application of dynamic_<>()cast

#include <iostream>
using namespace std;
class B
{ public:
B () {}
virtual void Function (){}
};
class D1 :public B
{ };
class D2 : public B
{ };
class D3: public B
{};
int main ()
{ B* Ab[3];
Ab[0] = new D1 ;
Ab[1] = new D2 ;
Ab[2] = new D3 ;
for ( int i =0; i<3; i++)
{D2 *ptrd2 = dynamic_cast<D2*>(Ab[i]);
if (ptrd2 != 0 )
cout << “D2 object found \n” ; }
return 0; }

The header file <typeinfo> also defines two functions bad_typeid and bad_cast. If the operator
typeid () is not successful it throws an exception bad_typeid. If dynamic_cast <>() is not
successful the bad_cast is thrown and it may lead to termination of the program.

THE STATIC_CAST<>()

The operator static_cast<>() performs conversion from one type to another in a way similar
to conventional C-type casting, i.e. (T) U except for the difference that in static_cast the
constantness of an object is not altered. The syntax is illustrated below.
static_cast<T>(object)
The expression converts the type of object to T. The type T may any one of the following.
(i) Fundamental type such as int, double, char, etc.
(ii) User defined types, i.e. class name.
(iii) Pointers.
(iv) References.
(v) The enum type – int and double may be cast to enum.
❖ 390 ❖ Programming with C++

As the name static implies the operator is applicable to static type checking and non-
polymorphic cases. The object of one class may be changed to object of another class provided
the appropriate construction functions are there. The following program provides an illustration
of application of static_cast <>() for pointers (from that of base class to derived class) and
references.

PROGRAM 15.15 – Illustrates application of static_cast<>()

#include <iostream>
using namespace std;
class Base
{ public :
int x;
int y;
void Speak () { cout << “I am class Base. \n” ;}
}; // end of Base class

class D : public Base

{ public:
void speak () { cout << “I am class D. \n”;}
Base B; // B is object of class Base.
int Area (Base B )
{return B.x*B.y;}
};
int main()
{
Base B ;
B.x =5;
B.y = 20;
Base * basep = &B; // pointer to base class.
cout<< “Function Speak() called by basep.”<<endl;
basep -> Speak();
cout<< “After static cast the function of D called.”<<endl;
D * dp = static_cast<D*>(basep); // basep cast to D type
dp->speak(); // Function speak() of D called by pointer dp
D* dp1 = (D*)basep; // C-type conversion
cout<<“After C-type conversion function call to speak().”<<endl;
dp1 ->speak();
D d; // d is declared an object of class D
cout <<“Area = “<< d.Area(B)<<endl; // call to function Area()
cout<< “After casting an object to reference by C-type conversion” << endl;
D &dp2 = (D&) B; // casting a reference by C-type casting
dp2.Speak(); // calling a function of class B
 Virtual Functions and Polymorphism ❖ 391❖
dp2.speak(); // calling a function of class D
D &dp3 = static_cast<D&>(B); //casting a reference by static cast
cout<< “After the static cast the functions are again called.”<< endl;
dp3.Speak(); //Speak() and speak() are Two different functions.
dp3.speak();
return 0 ;
}

The output is given below. The comments given above have made the output self explanatory.
Function Speak() called by basep.
I am class Base.
After static cast the function of D called.
I am class D.
After C-type conversion function call to speak().
I am class D.
Area = 100
After casting an object to reference type by C-type conversion
I am class Base.
I am class D.
After the static cast the functions are again called.
I am class Base.
I am class D.

THE REINTERPRET_CAST <>()

The operator is coded as below.
The reinterpret_cast <T>(object)
The operator returns a reinterpretation of its operand which is quite different from that of
the operand. For instance, it can cast an int into a pointer type and a pointer into an int type or
change a pointer to an integer into a pointer to char, etc. Both the dynamic cast and
reinterpret_cast are prone to errors and should be used with caution. Some compilers also give
a warning to this aspect.
The reinterpret_cast can be used on pointers as well as references. It is applicable to static
as well as to dynamic objects. Like static_cast it also preserves the constantness of an object. The
following program illustrates the application of the reinterpret_cast for conversion of pointers
and references. The operator can be applied to objects of incomplete classes as well without getting
an error message from compiler. Some of these applications are illustrated in the following
program.

PROGRAM 15.16 – Illustrates application of reinterpret_cast<>()

#include <iostream>
#include <string>
using namespace std;
❖ 392 ❖ Programming with C++
class A; // incomplete class
class B; // incomplete class
class Base
{ public :
void Speak () { cout << “ I am class Base. \n” ;}
};
class D : public Base
{ public:
void speak () { cout << “ I am class D. \n”;}
};

int main()
{ int x = 65;
int * px = &x; // px is pointer to int x
char* chp = (char*) px; //px converted into pointer char* chp
// by C-type casting
cout << “*chp = ” << *chp << endl; // output statement for *chp
cout << *px <<endl;
// below similar conversion done by reinterpret_cast<>()

char* ptrx = reinterpret_cast<char*> (px);

cout <<“*ptrx = ” << *ptrx<<endl;

D* ptrd = reinterpret_cast< D*>( px); //px converted into type D
ptrd ->speak();
B * bp ; // It also applies to incomplete classes
A* ap = reinterpret_cast<A*>(bp);
}

The expected output is given below. The output demonstrates the successful conversion of the
pointers.
*chp = A
65
*ptrx = A
I am class D.

In the following program the reinterpret_cast <>() is used with pointers to functions.

PROGRAM 15.17 – Illustrates application of reinterpret_cast <>() with functions

#include <iostream>
using namespace std;
void Funct1 ( )
{ cout<< “This is void function with void argument \n”;}
void Funct2(int m, int n)
{ cout<<“This is void function with two arguments.\n”;}
 Virtual Functions and Polymorphism ❖ 393❖
int Funct3 (int x, int y )
{return x*y;}
int main()
{ int x = 65;
int * px = &x; // px is pointer to int x
typedef void (*PF)( ); // typedef of pointers to void functions
PF function1 = (PF)Funct1; // conventional C-type cast
function1() ; // function call
PF function2 = reinterpret_cast<PF>(Funct1);//by reinterpret_cast
function2 (); // function2 called
int y = 2;
PF Function2 = reinterpret_cast<PF>(Funct2);
Function2();
typedef int(*pF)(int, int); //type definition of pointers to
// int functions having 2 int parameters
pF function3 = (pF) Funct3 ; // conventional C-type cast
cout << function3(x,y )<<endl; // function call
//below similar conversion is done by reinterpret_cast<>()
pF function4 = reinterpret_cast<pF>(Funct3);
cout << function4 (x,y)<< endl;
return 0;
}

The output is given below.

This is void function with void argument
This is void function with void argument
This is void function with two arguments.
130
130
In the above out put, the first two lines are due to function calls function1() and function
2(). The third line of output is due to function call of function2(x,y). The pointer of funct3 is
again converted to pointer of function3(x,y) and into pointer to function4 (x,y). The last two lines
of out put are due to call of these two functions.

THE CONST_CAST<>()

The operator provides an access to a variable with the attributes const or volatile. The objects
declared volatile may be modified from outside the control of the program. The operator
const_cast<>()does not change the basic attribute of the variable. The syntax of the operator is
as follows.
const_cast <T*> ( ptr_object)
The following program tries to bring out the action of const_cast.
❖ 394 ❖ Programming with C++

PROGRAM 15.18 – Illustrates the application of const_cast<> ().

#include <iostream>
using namespace std;
void F ( char* M )
{ cout<< M <<endl;}
int main()
{ const char * S= “Delhi”;
//F (S); Error, cannot convert from const char* to char*
F ( const_cast<char*> (S)); // it is alright

const int n = 10;

const int *ptrn = &n;
// n= n+5; Error, the n is const. But the following is alright

*(const_cast<int*>(ptrn) ) = n + 5;
cout <<*(const_cast<int*>(ptrn) )<<endl;

//const_cast<int>(n) = n+8; This gives error, const_cast

//cannot convert from const int to int.
cout << n <<endl;
return 0 ;
}

The expected output is given below.

Delhi
15
10

EXERCISES
1. How do you declare pointer to a class?
2. How do you make the base class pointer point to object of derived class?
3. What is the benefit of declaring a base class function a virtual function?
4. What is a virtual base class?
5. What is the difference between a virtual function and an overloaded function?
6. What are abstract classes? How are they useful?
7. What do you understand by the term dynamic binding?
8. What do you understand by polymorphism?
9. Why constructors are not declared virtual while the destructors are declared virtual?
10. What is the benefit of declaring destructors virtual?
11. Make a program for illustrating pure virtual function Speak() in the base class Animal. The derived
classes are class Dog, class Cat, class EMan for Englishman and class HMan for Indian. The
function Speak() displays their respective mother tongue.
 Virtual Functions and Polymorphism ❖ 395❖
Answer:
PROGRAM 15.19 – Another example of abstract class and pure virtual function.
#include <iostream>
using namespace std;
class Animal {
public:

virtual void Speak() =0;

};

class Dog : public Animal

{ public:
void Speak ( )
{cout<<“ BHOO BHOO” <<endl;}
};
class Cat : public Animal
{
public:
void Speak (void)
{cout<< “ Miewn Miewn” << endl;;}
};
class EMan : public Animal
{ public :
void Speak(){ cout << “I speak English.”<<endl;}
};
class HMan : public Animal
{ public :
void Speak(){ cout << “I speak Hindi.”<<endl;}
};
void main ()
{
Dog D ;
Cat C ;
EMan EM;
HMan HM;

D.Speak();
C.Speak();
EM.Speak();
HM.Speak();
cout<<“\n”;
}

The expected output is given below.

BHOO BHOO
Miewn Miewn
I speak English.
I speak Hindi.
❍❍❍
❖ 396 ❖ Programming with C++

CHAPTER

16.1 INTRODUCTION

There are a large number of programs which carry out similar operations on different types of
data. For example, sorting of lists, whether the list consists of integers or floating point numbers
or characters the operational code is similar. Similarly in swap of two quantities the code is
identical except for the type of variables. Same is true in the case of finding the greater of two
quantities whether they are integers, floating decimal point numbers or characters. In C++
whenever a variable is declared, its type has to be mentioned before its name. Therefore, even
though the codes are similar, for making them applicable to different data types, i.e.
int, double or char, etc., the programmer has to write a number of programs one for each
data type. Templates give the facility that you need not write a number of programs. With
template we need write only one program that is equally applicable to all types of data — the
fundamental types as well as user defined types. In case of templates the program is made with
a place holder in place of type. When the type is substituted or referred to, the compiler
substitutes the specified type in place of the place holder and compiles the program. Now if
another type is specified, the compiler generates another template specialization for that type and
compiles the program. So template is like a mould which allows different types to be cast for
generating the corresponding programs.
Template programming is also called generic programming. The function templates and class
templates are also called generic functions and generic classes respectively. There are other methods
also to achieve the same end such as use of macros, # define and void pointers, etc. Such programs
were used by programmers before the implementation of templates, however, these methods have
disadvantages which are not there in templates. For instance, macros do not allow strict type
checking while the templates do.
Use of templates started around 1991. The greatest advantage of generic programming is that
it is applicable to any type of data, i.e. the fundamental types as well as to user defined types.
In the following we first discuss the functions templates and then the class templates.

16.2 FUNCTION TEMPLATES

A function template declaration in which the function and its parameters are of same type is
illustrated below. A function template is declared with keyword template followed by angular
 Templates ❖ 397❖
brackets which contain the keyword typename or class followed by an identifier or name of the
typename or class, i.e. the name of place holder for the type. After the angular brackets the function
is defined. See illustration below.
template <typename type_identifier > type_identifier function_name (type_identifier
parameter1, – , –)
{ statements; }
Or it may be coded with keyword class in place of typename as below.
template < class type_identifier > type_identifier function_name (type_identifier
parameter1, – , – )
{ statements; }
In the above declarations the words template, typename and class are keywords of C++. Both
the above declarations are equivalent. A type_identifier is any valid identifier in C++ which is used
as a place holder for type. The type_identifier used for class or typename has to be placed before
the function name as well as before the name of every parameter of the function. For instance,
let T be the type_identifier or name of the typename or class, a function template for calculation
of area of a rectangle with sides represented by x and y may be defined as given below.
template <typename T> T Area (T x, T y) // function head
{ return x * y ; } // function body
Here T is the type_identifier. The T placed before the name area is its return type. The above
function may also be defined with the keyword class as below.
template < class T > T Area ( T x, T y )
{return x * y;}
Both the function definitions given above are equivalent and any one may be used. Here T
is the type_identifier, in fact, it is a place holder for the type which is still not specified and will
be substituted later. Area is the name of function. The sides x and y are the two parameters of
function Area (). The variable x may be length of rectangle and y may be width of rectangle and
Area is the area of rectangle. Note that the name T appears before Area as well as before x and
before y in the head of the function. The first line in the above definition may be written in two
lines as illustrated in program 16.3.

CALLING A FUNCTION TEMPLATE

The function template is called by mentioning the name of function followed by the actual type
in angular brackets, followed by list of arguments in parentheses ( ). It is illustrated below.
Function_name <type> (arguments);
An example of calling template function is given below.
int x = a, y = b; // a and b are arguments
Area<int>(a,b); // calling the function Area
Now if the sides and Area are double numbers, the function call would become,
double x = A, y = B ; //A and B are arguments
Area <double> (a, b); // function call
❖ 398 ❖ Programming with C++

The following program illustrates the use of function template which calculates area of
rectangle whose sides are either in integers or double numbers.

PROGRAM 16.1 – Illustrates definition and calling of a function template.

#include <iostream>
using namespace std;
template <typename T > T Area (T x, T y ) //function template
{return x*y ; } // definition
int main()
{
int n = 4 , m = 3;
double j = 6.5 , k = 4.1 ;
cout<<“Area1 = ”<<Area<int>(n, m)<<“\t Area2 = ” << Area <double> (j,k) <<
endl;
return 0;
}

The expected output is given below. Area1 represents the area for rectangle with sides (4, 3)
and Area2 represents the area for rectangle with sides 6.5 and 4.1.
Area1 = 12 Area2 = 26.65
In the above program T is in fact place holder for the type. The template program is compiled
only when the actual type of variables is substituted for T. In Program 16.1, in
the call Area<int>(n,m) the type of variable area is int and in the call Area
<double>(j,k)the type of return value is double. Both types of return values, i.e. int and
double are obtained with the same function definition.
Another example is provided in Program 16.2. In this program we have not mentioned the
type in the angular brackets at the time of function call. The compiler, in fact, can know the type
of variables it is dealing with because the type is mentioned at the time of declaration of the
variables. Even if the type <int> or <double> is not included in the function call the compiler
can find out the type from declarations of variables, for instance, when it encounters integers like
int m, n; the T is replaced by int and the program is generated for integers. When it comes
across floating point numbers declared as double j, k ; in the program the compiler regenerates
the template specialization for double numbers. Moreover, you may use any valid identifier
(name) in place of T. However, it is better to explicitly mention the type of return value,
particularly, when the function template contains more than one type of parameters.
Below we define a function template for determining the maximum of two numbers with
the declaration <class T> instead of <typename T>. The variables may be integers, floating
point numbers or characters. Characters also have integral values as per ASCII code. For instance
‘A’ = 65 and ‘B’ = 66 and so on. The function template definition is given below. This function
determines greater of the two numbers.
template <class T> T max(T x, T y)
{ return x > y ? x:y ;}
 Templates ❖ 399❖
In case a template function is defined below the main (), the template prototype or template
declaration of function should be given above main (). As explained above for Program 16.1,
T is the identifier or name for type, max is the name of function, x and y are the two parameters
of max(). Note that max is preceded by identifier T and the two parameters are also individually
preceded by identifier T. This is very similar to the usual declaration of function except for the
type which is now represented by T. The following program illustrates the declaration of a function
template which finds the greater of the two variables which may be of type int or double or char.

PROGRAM 16.2 – Illustrates function template for finding greater of two numbers.

#include <iostream>
using namespace std;
template <class T> T max(T x, T y) // function definition
{return x>y ? x : y;} // if(x>y) return x else return y
int main()
{ int n = 64 , m = 67 ; // int variables
char ch1 = ‘S’ , ch2 = ‘R’; // char variables
double D1 = 6.87, D2= 8.34; // double variable
cout << “max (n,m)= ”<<max <int>(n,m)<<endl;
cout<< “max (ch1,ch2) = ” << max <char>(ch1,ch2)<<endl;
cout<< “max(D1,D2) = ” << max<double>( D1, D2)<< endl;
return 0;
}

The expected output is given below. The program and the results are self explanatory.
max (n,m)= 67
max (ch1,ch2) = S
max(D1,D2) = 8.34
The following program makes a function template for Swap function which exchanges the
values of two variables. The quantities involved may be integers, double numbers, characters or
strings. Strings are objects of class string (C++ strings) and hence a user defined type. Same
function template can be used for all the types of data. You should note that Swap() is user
defined function and is different from swap() of C++ Standard Library because S is in this is in
upper case.

PROGRAM 16.3 – Illustrate a function template for swapping two quantities.

#include <iostream>
# include <string>
using namespace std;
template <class T> // Template definition of Swap()
void Swap (T &x , T &y) // values passed on by reference
{ T temp;
temp = x;
❖ 400 ❖ Programming with C++
x = y;
y = temp;}

void main()
{ int n =60, m=30;
char ch1= ‘A’, ch2 = ‘D’;
double a = 2.45 , b = 4.76;

string S1 = “Morning”; //S1 and S2 are string objects

string S2 = “Evening”; // their type is string

Swap <double>(a,b);
Swap<char>(ch1,ch2);
Swap<int>(n,m);
Swap<string>(S1,S2 ); // swapping user defined data

cout <<“a = ”<<a<<“\tb = ”<<b<<“\n”;

cout<< “ch1 = ”<<ch1<<“\tch2 = ” << ch2<< endl;
cout <<“n = ”<<n <<“ \tm = ”<<m<<“\n”;
cout << “S1 = ”<< S1 << “ , S2 = ”<< S2<< endl;
}

The expected output is given below.

a = 4.76 b = 2.45
ch1 = D ch2 = A
n = 30 m = 60
S1 = Evening , S2 = Morning
Before the implementation of Swap function, the initial values are a = 2.45, b = 4.76, ch1
=A, ch2 = D, n = 60, m = 30 and string S1 = “Morning” and S2 = “Evening”. The values have
been interchanged by implementing Swap function. Also note that the variables are passed by
reference in the declaration of Swap function. As we have already discussed earlier (in Chapter
7 on functions and in Chapter 9 on pointers and references) in passing variables by values, only
copies of variable values are passed on and variables do not get changed. For a change in the values
of variables they have to be passed on to functions by references or pointers.

16.3 FUNCTION TEMPLATE WITH ARRAY AS A PARAMETER

The following program illustrates a function template for output of an array. The function
template Display () is declared as below.
template <class T>
void Display(const T*A, unsigned int size)
Here size stands for number of elements in the array, T is the name of place holder for type
and A is the name of array which is a constant pointer to the first element of array. The
application of code is illustrated in the following program,
 Templates ❖ 401❖
PROGRAM 16.4 – Illustrates a function template for output of an array.

#include <iostream>
using namespace std;
template <class T>
void Display(const T*A, unsigned int size)// function declaration
{ for( int i =0 ; i<size; i++ ) // and definition
cout<< A[i] <<“ ”;
cout<<endl;}
int main ()
{const int K = 6,M =15, J = 4 ;
int Bill[J] = {20, 30, 40, 50}; // an array of integers
// Below is an array of doubles
double Kim[K] = { 4.5, 5.6, 7.8, 8.9, 1.2, 3.4};
char Name[M] = “B.L.Juneja”; // a string of characters
Display (Bill, J);
Display (Kim , K);
Display (Name , M);
return 0;
}

The expected output is given below.

20 30 40 50
4.5 5.6 7.8 8.9 1.2 3.4
B.L.Juneja
Below the Program 16.5 is extended to include strings along with characters, int and double
numbers. The same function template Display() is used to display on the monitor an array of
integers, an array of double numbers, an array of characters and an array of strings.

PROGRAM 16.5 – Illustrates a function template for displaying arrays and strings.

#include <iostream>
using namespace std;
template <class T>
void Display(T A[],unsigned int size)
{ for ( int i =0 ; i<size;i++ )
cout<< A[i] <<“ ”;
cout<<endl;}
int main ()
{
const int K = 6,M =5, J = 7, N = 4 ;
❖ 402 ❖ Programming with C++
int Bill[J] = {20,30, 40,50,60,70,80};
char ch [N] = { ‘D’, ‘E’, ‘X’, ‘S’};
double Kim[K] = { 4.5, 5.6, 7.8, 8.9 , 1.2 , 3.2};
char* S[M] ={“Malini” , “Sneha” , “Sunita” , “Pushkar”, “Alka”};
Display( Bill , J); // calling template function
Display (Kim , K);
Display (ch,N);
Display ( S, M);
return 0;
}

The output is as under.

20 30 40 50 60 70 80
4.5 5.6 7.8 8.9 1.2 3.2
DEXS
Malini Sneha Sunita Pushkar Alka

Sorting of lists is very common in programming. The list may be of elements of type int,
doubles or characters, the codes for sorting are similar. Therefore, in the following program a
template function is developed for sorting the elements of an array in ascending order.

PROGRAM 16.6 – Illustrates a function template for sorting of lists.

#include <iostream>
using namespace std;
template <class T> // sorting function template
void Listsort (T A[], const int n)
{for ( int i = 0 ; i<n;i++)
{for(int j =0;j<n–1;j++)
if(A[j]>A[j+1])

swap( A[j],A[j+1]);}}

template <class T>

void Display( T A[] , unsigned int size)

{ for ( int i =0 ; i<size; i++ )

cout<< A[i] <<“ ”;
cout<<endl;}

int main ()
{ const int K = 6,M =4, J = 7, N = 4 ;
int Bill[J] = {20,30,70, 80,40,50,60};
 Templates ❖ 403❖
char ch [5] = { ‘D’, ‘E’, ‘X’, ‘S’, ‘Z’};
double Kim[K] = { 4.5, 5.6, 1.2 ,7.8, 8.9 , 3.2};
Listsort(Bill,J);
Display( Bill , J);
Listsort(Kim,K);
Display <double>(Kim , K);
Listsort (ch,5);
Display<char> (ch,5);
return 0;
}

The expected output is given below.

20 30 40 50 60 70 80
1.2 3.2 4.5 5.6 7.8 8.9
DESXZ

TEMPLATE FUNCTION FOR SWAPPING SPECIFIED NUMBER OF ELEMENTS OF ARRAYS

In Program 16.7 given below a Swap template function is developed which swaps the specified
number of elements of two arrays which may be of type int, double or character etc. The name
genric has been used as the class name.

PROGRAM 16.7 – Illustrates a function template for swapping a specified number of

elements of two arrays.

#include <iostream>
using namespace std;
template <class genric> // declaration and definition of Swap
void Swap ( genric x[ ], genric y[ ],int m )
{
genric temp ; // type identifier is genric in place of T
for (int i = 0; i<m;i++)
{temp = x[i];
x[i] = y[i];
y[i] = temp ; }
} // end of template function
void main()
{ double Bill[ ] = { 11.1, 12.2, 13.3,14.4 ,15.5};
double Kim [] = { 30.5, 31.1, 32.2, 33.3, 34.4};
char Ch[ ] = { ‘A’, ‘B’, ‘C’,’D’,’E’ };
char Kh [] = { ‘S’, ‘T’,’U’,’V’,’X’};
❖ 404 ❖ Programming with C++
int K[5] = {1 ,2 , 3, 4,5};
int M[5] = {50,60,70,80,90};
cout << “Before swap the arrays are as below\n”;
cout <<“Bill\tKim\tCh\tKh\tK\t M ” << endl;
for ( int k = 0; k <5; k++)
cout <<Bill[k] <<“\t”<<Kim[k]<<“\t”<<Ch[k]<<“\t”<<Kh[k] <<“\t” <<
K[k]<<“\t”<<M[k] <<endl;
Swap (Bill, Kim,1 );
Swap ( K , M ,3 );
Swap ( Ch, Kh,5 ) ;
cout << “After swapping the arrays are \n”;
cout << “ Bill\tKim\tCh\tKh\tK\tM ” << endl;
for ( int j = 0; j <5; j++)
cout <<Bill[j] <<“\t”<<Kim[j]<<“\t”<<Ch[j]<<“\t”<<Kh[j]<< “\t”<<
K[j]<<“\t”<<M[j]<<endl;
}

The expected output is given below.

Before swap the arrays are as below
Bill Kim Ch Kh K M
11.1 30.5 A S 1 50
12.2 31.1 B T 2 60
13.3 32.2 C U 3 70
14.4 33.3 D V 4 80
15.5 34.4 E X 5 90
After swapping the arrays are
Bill Kim Ch Kh K M
30.5 11.1 S A 50 1
12.2 31.1 T B 60 2
13.3 32.2 U C 70 3
14.4 33.3 V D 4 80
15.5 34.4 X E 5 90

As specified in the program between Bill and Kim only one element has been exchanged,
between Ch and Kh all the five elements have been exchanged and between K and M only 3
elements have been exchanged as instructed in the code.

16.4 FUNCTION TEMPLATES WITH MULTIPLE TYPE ARGUMENTS

Up till now we have discussed function templates which involved only one type of variables. For
instance, all the function parameters are either integers only or only double numbers or only char,
etc., and return values are also of same type. But there are ample situations wherein the functions
 Templates ❖ 405❖
parameters are of more than one type like int and double and you may like to have the return
value either in integers or double number. The illustrations of template declarations for such
cases are given below. The template function selects greater of the two values which may not be
of same type. Therefore, two place holders are declared for class name.
template <class T, class E >
T max ( T x , E y)
{return x > y ? x : y ; }
In this declaration the output will be of type T because we have taken max of type T. If it
is desired to have output of type E we may declare the above expression as below.
template< class T, class E>
E max ( T x , E y)
{return x > y ? x : y ;)}
The following program gives an illustration of above function template in finding out the
greater of two values whether they are of type char, int or a double, etc.

PROGRAM 16.8 – Illustrate functions template for more than one data types.

#include <iostream>
using namespace std;
template <class T , class E> E max(T x, E y)
{return x>y ? x : y;}

int main()
{
int n = 88 ;
double m = 80.4 ;
int ch1 = 67 ;
char ch2 = ‘A’;
double D1 = 76.87 ,D2= 90.3;
cout << “max (n, m)= ”<<max (n,m)<<endl;
cout<< “max (ch1,ch2) = ” << max (ch1,ch2)<<endl;
cout<< “max (D1,D2) = ” << max ( D1, D2)<< endl;
cout<< “max (n,D2) = ” << max (n,D2)<<endl;
return 0;
}

The expected output is given below

max (n, m)= 88
max (ch1,ch2) = C
max (D1,D2) = 90.3
max (n,D2) =90.3
❖ 406 ❖ Programming with C++

From the output it is clear that the output is of the type of the second variable in the function
because we specified it so in the function declaration, i.e. E max(T x, E y). If the function
head is declared as given below,
T max(T x, E y)
the output would be of type T.

16.5 OVERLOADING OF TEMPLATE FUNCTIONS

The template function overloading is similar to non-template function overloading. In

overloaded function the function name is same while the arguments are different in number and
or in type. The following program illustrates overloaded template functions.

PROGRAM 16.9 – Illustrates overloading of template functions with parameters of single

type.

#include <iostream>
using namespace std;
template <class T>
T Product (T x, T y)
{return x*y;}

template <class T>

T Product ( T x, T y, T z)
{return x*y*z;}
int main ()
{ int a (2), b(3), c(4);
double A(2.0), B(5.5), C(1.5);
cout<<“Product (A,B,C) = ”<< Product (A,B,C)<< endl;;
cout<<“Product ( A,B ) = ”<< Product ( A,B )<< endl;
cout << “Product ( a,b,c ) = ”<< Product ( a,b,c)<<endl;
return 0;
}

The expected output is given below.

Product (A,B,C) = 16.5
Product ( A,B ) = 11
Product ( a,b,c ) = 24
Every function in the above program has one type of variable. It is either int, or double. In
the following, we take the functions with multiple type arguments.

PROGRAM 16.10 – Illustrates overloaded template functions with parameters of multiple

types.
#include <iostream>
using namespace std;
 Templates ❖ 407❖
template <class T, class U>
U Product ( T x, U y)
{return x*y;}
template <class T, class S, class U>
S Product ( T x, S y, U z)
{return x*y*z;}
int main ()
{ int a (1);
double B(1.5);
char Ch(‘D’);
cout<<“Product (a,B,Ch) = “<< Product (a,B,Ch)<< endl;
cout<<“Product (a,B) = “<< Product (a,B)<< endl;
cout<<“Product( ‘A’, 4.5, 6) = “ <<Product ( ‘A’, 4.5, 6)<< endl;
return 0;
}

The expected output is given below.

Product (a,B,Ch) = 102
Product (a,B) = 1.5
Product ( ‘A’, 4.5, 6) = 1755

16.6 CLASS TEMPLATES

We have already got some experience with function templates. Different data types may be used
for the evaluation of the function. The class template has similar properties. Class objects having
different types of data can use the same class template. A class template is declared as below.
template < class type_identifier > class identifier_for_class
For example the class template Cuboid of the following program is declared as below.
template <class T> class Cuboid
Here Cuboid is the name of class while T is the name of place holder for type. You may use
any valid name in place of T. Also wherever the variables are declared in the class, they would
be preceded by type identifier T. The class object is declared as below.
Class_name <type> object_name;
The following program is a class template for the class Cuboid which calculates the surface
area and volume of rectangular prismatic bodies. The dimensions may be in int or float, double,
long or short. In this program the member functions are defined inside the class.

PROGRAM 16.11 – Template class with member functions defined inside the class.

#include <iostream>
using namespace std;
template <class T>
❖ 408 ❖ Programming with C++
class Cuboid {
T x , y , z ; // The variables x , y, z are of same type
public:
Cuboid(T L, T W, T H){x = L; y = W; z = H ;}

T surface_area( )
{return 2*(x*y +y*z +z*x);

T volume( ) { return x*y*z ;}

};
// end of class
int main()
{Cuboid <int> C1(3,8,5); // dimensions in int
Cuboid < double> C2(3.5,5.5,4.5) ; // dimensions in doubles

cout << “Volume of C1= ”<< C1.volume()<<“\n”;

cout<< “Volume of C2 = ” << C2.volume()<<“\n”;

cout<<“surface area of C1 = ”<< C1.surface_area()<<“\n”;

cout<<“surface area of C2 = ”<< C2.surface_area()<<“\n”;
return 0 ;
}

The expected output is given below.

Volume of C1= 120
Volume of C2 = 86.625
surface area of C1 = 158
surface area of C2 = 119.5

DEFINING FUNCTION TEMPLATE OUTSIDE THE CLASS TEMPLATE

In the above program all the function are defined inside the class template. The code for defining
a function template outside the class template is illustrated below.
T class_identifier <T> :: function_identifier ( type_identifier parameter1, .. )
{ statements ;}
An illustration of this, the declaration of function volume class cuboid is given below. An
application is illustrated in Program 16.12.
template <class T>
T Cuboid <T> ::volume()
{return x*y*z ;}

PROGRAM 16.12 – Illustrates definition of template member functions outside the class
template. Constructor is defined inside the class.

#include <iostream>
using namespace std;
template <class T> // Declaration of class template
 Templates ❖ 409❖
class Cuboid {
public:
Tx,y,z;
Cuboid (T L, T W, T H){x = L; y = W; z = H ;} // constructor
T surface_area( ); // function prototype in the class body
T volume( ); // function prototype in the class body
}; // end of class
template <class T > // definition of function surface area
T Cuboid <T>::surface_area()
{return 2*(x*y +y*z +z*x);}
template <class T > // definition of function volume
T Cuboid <T>::volume()
{return x*y*z ;}
int main()
{Cuboid <int> C1(5,6,4); // Object with int dimensions
Cuboid <double> C2(2.2,3.5,4.5) ; // An object with dimensions
// in floating point numbers.
cout << “Volume of C1= ” << C1.volume()<<“\n”;
cout<< “Volume of C2 = ” << C2.volume()<<“\n”;
cout<<“surface area of C1 = ” << C1.surface_area()<<“\n”;
cout<<“surface area of C2 = ” << C2.surface_area()<<“\n”;
return 0 ;
}

The expected output of the above program is given below.

Volume of C1= 120
Volume of C2 = 34.65
surface area of C1 = 148
surface area of C2 = 66.7
The class object C1 has integer type data while object C2 has data of type double. The class
template processes both types of data.
In the following program dealing with class template, all the function prototypes including
the constructor function are declared in the class template and defined outside the class template.

PROGRAM 16.13 – Illustrates class template with all functions defined outside the class.

#include <iostream>
using namespace std;
template <class T>
class Cuboid {
public:
❖ 410 ❖ Programming with C++
Cuboid (T , T, T ); // constructor prototype
T surface_area( ); // prototype of surface_area ()
T volume( ); // prototype of volume () function
private:
T x , y, z;
};
template < class T> // definition of constructor function
Cuboid <T>::Cuboid <T> ( T L, T W, T H ) { x = L; y= W; z=H ;}
template < class T> // definition of surface_area ()
T Cuboid <T>::surface_area()
{ return 2*(x*y +y*z +z*x);}
template < class T> // definition of function volume
T Cuboid <T>::volume()
{ return x*y*z ;}
int main()
{
Cuboid <int> C1(5,6,4);
Cuboid <double> C2(2.5,3.5,4.5) ;
cout << “Volume of C1= ” << C1.volume()<<“\n”;
cout << “Volume of C2= ” << C2.volume()<<“\n”;
cout << “surface area of C1 = ”<< C1.surface_area()<<“\n”;
cout << “surface area of C2 = ”<< C2.surface_area()<<“\n”;
return 0 ;
}

The expected output of the program is given below.

Volume of C1= 120
Volume of C2= 39.375
surface area of C1 = 148
surface area of C2 = 71.5

16.7 FRIEND FUNCTION TEMPLATE TO CLASS TEMPLATE

The declaration of friend function template in the class is similar to the non-template function
declaration. The definition of friend function is done outside the class without the scope
resolution operator.
The following program illustrated the declaration of a friend function template with name
Area in a class template called Rect which deals with rectangles. The class template includes private
data i.e., sides of rectangle, constructor function and declaration of friend function which is
defined outside the class. The class object is the argument of the friend function.
 Templates ❖ 411❖
PROGRAM 16.14 – Illustrates friend function template to class template.

#include <iostream>
using namespace std;

template <class T>

class Rect {
friend T Area (const Rect &R); //declaration of friend function
// template
private :
T x; // x and y are sides of rectangle
T y;
public :
Rect (T A, T B) { x = A , y = B; } // constructor
}; // end of class template
template <class T> // definition of friend function
T Area( const Rect <T> &R){ return R.x* R.y ;} // R is an object
int main ()
{
Rect<int> myrect(6, 2); // myrect is an object of Rect
cout<<“Area = ” << Area(myrect) << endl;;
return 0;
}

The expected output is given below.

Area = 12

PROGRAM 16.15 – Illustrates friend function template to a class template with data of
multiple types.

#include <iostream>
using namespace std;
template <class T, class U> //class template with multiple types
class Rect
{
private :
T x;
U y;
public :
Rect (T A, U B) { x = A , y = B; }
friend U Area (const Rect &R); // Friend function template
};
#include <iostream>
❖ 412 ❖ Programming with C++
using namespace std;
template <class T, class U>
U Product ( T x, U y)
{return x*y;}
template <class T, class S, class U>
S Product ( T x, S y, U z)
{return x*y*z;}
int main ()
{ int a (1);
double B(1.5);
char Ch(‘D’);
cout<<“Product (a,B,Ch) = ”<< Product (a,B,Ch)<< endl;;
cout<<“Product (a,B) = ”<< Product (a,B)<< endl;
cout <<“Product ( ‘A’, 4.5, 6) = “ <<Product (‘A’, 4.5, 6)<<endl;
return 0;
}

The expected output is given below.

Product (a,B,Ch) = 102
Product (a,B) = 1.5
Product ( ‘A’, 4.5, 6) = 1755

The following program is another example of a friend function template to a class template.

PROGRAM 16.16 – Illustrates another application of friend function template to class a

template.

#include <iostream>
using namespace std;
template <class T> // type declaration
class Cuboid { // beginning of class template
private:
T x , y, z;
public:
Cuboid ( T, T, T); // constructor prototype
T surface_area( );
T volume( );
// Friend function template
friend T costpaint ( T, const Cuboid <T> &C);
}; //end of class Cuboid
 Templates ❖ 413❖
template < class T> // definition of constructor outside class
Cuboid <T> :: Cuboid<T>( T L, T W, T H) { x = L ;y = W ; z =H ;}
template <class T> // definition of function surface area
T Cuboid <T>::surface_area()
{return 2*(x*y +y*z +z*x);}
template <class T> // definition of volume
T Cuboid <T>::volume()
{return x*y*z ;}
//Below is the definition of friend function template
template <class T >
T costpaint (T Rho, Cuboid<T> &C )
{return Rho * C.surface_area() ;}
int main()
{Cuboid <int> C1 (4, 5, 6 ) ;
Cuboid <double> C2 ( 2.5, 3.0, 4.0 );
int D1 = 3 ;
double D2 = 2.0;
cout << “Volume of C1 = ” <<C1.volume()<<“\n”;
cout<< “Volume of C2 = ” <<C2.volume()<<“\n”;
cout<<“surface area of C1 = ” <<C1.surface_area()<<“\n”;
cout<<“surface area of C2 = ” <<C2.surface_area()<<“\n”;
cout <<“cost of painting P1 = ” <<costpaint ( D1 ,C1 )<<“\n”;
cout<<“cost of painting P2 = ” <<costpaint (D2 , C2 )<<“\n”;
return 0 ;
}

The expected output is given below.

Volume of C1 = 120
Volume of C2 = 30
surface area of C1 = 148
surface area of C2 = 59
cost of painting P1 = 444
cost of painting P2 = 118

16.8 FRIEND CLASS TEMPLATE TO A CLASS TEMPLATE

The process of declaration of friend class template is similar to that for non template classes
except for the type of variables involved. It is illustrated in the following program in which a
friend class template is declared by name paint. The class paint is declared as friend to class
Cuboid. The declaration of the class paint is illustrated below.
❖ 414 ❖ Programming with C++
template <class U > class paint ;
In the body of class Cuboid, the class paint is declared as
friend class paint;

PROGRAM 16.17 – Illustrates friend class template to a class template.

template <class U> class paint;

#include <iostream>
using namespace std;
template <class T>
class Cuboid {
friend class paint; // declaration of class paint as friend
public:
Cuboid ( T, T, T); // constructor protoype of class Cuboid
T surface_area(); //Prototypes of area() and volume()
T volume();
private:
T x, y, z; // x , y and z are of same type
}; // End of class Cuboid
template < class T> // constructor definition for class Cuboid
Cuboid <T> :: Cuboid<T>( T L, T W, T H) { x = L ;y = W ; z =H ;}
template <class T> // definition of function area()
T Cuboid <T>::surface_area()
{return 2*(x*y +y*z +z*x);}
template <class T> // Definition of function volume
T Cuboid <T>::volume()
{return x*y*z ;}
template <class U> // class paint
class paint { //definition of class paint
private:
U Rho;
public:
paint (U); // prototype of constructor
paint(){Rho =2;}; // default constructor
U cost( U Rho,U surface_area ) // Definition of cost()
{
U CP; // CP = cost of paint
CP = Rho* surface_area ;
return CP;}
}; // end of class paint
 Templates ❖ 415❖

template < class U> // constructor for class paint.

paint<U>:: paint<U>( U D ) { Rho = D;}
int main()
{ Cuboid <int> C1 (4, 5, 6 ) ;
Cuboid <int> C2 ( 2, 3, 4);
paint <int> P1;
paint <int> P2;
int R1 =2;
int R2 = 3;
cout << “Volume of C1 = ” <<C1.volume()<<“\n”;
cout<< “Volume of C2 = ” <<C2.volume()<<“\n”;
cout<<“surface area of C1 = ”<<C1.surface_area()<<“\n”;
cout<<“surface area of C2 = ”<<C2.surface_area()<<“\n”;
cout <<“cost of painting P1 = ”<<P1.cost(7 ,C1.surface_area())<< “\n”;
cout<<“cost of painting P2 = ”<< P2.cost (7 , C2.surface_area() )<<“\n”;
return 0 ;
}

The expected output is given below.

Volume of C1 = 120
Volume of C2 = 24
surface area of C1 = 148
surface area of C2 = 52
cost of painting P1 = 1036
cost of painting P2 = 364

16.9 TEMPLATE CLASS FOR COMPLEX VARIABLES

The header file <complex> contains the template class for manipulation of complex type variables.
The following program illustrates some arithmetic operation on complex variables by including
this file in the program. A variable consists of a pair of values, i.e. the real and the imaginary
values.

PROGRAM 16.18 – Illustrates manipulation of complex variables.

#include<iostream>
#include<complex>
using namespace std;
int main ()
{ complex <double> a(4.5, 5.5) ;
complex <int> b(2,3) ;
complex <int> g(3,4) ;
complex<double> e (1.4, 3.4);
complex <double> C = a + e ;
❖ 416 ❖ Programming with C++
complex <int>D = g - b;
complex<int> Product = b*b ;
cout<<“a = “ <<a<<“\t”<<“b = “<<b <<“\n”;
cout<<“C= “<<C<<“\t”<<“D = “<<D <<“\n”;
cout << “Product b*b = “<< Product<<endl;
return 0;
}.

The expected output is given below

a = (4.5,5.5) b = (2,3)
C= (5.9,8.9) D = (1,1)
Product b*b = (-5,12)

EXERCISES
1. What is a function template?
2. What is a class template?
3. What do you understand by generic programming?
4. How do you declare a function template?
5. How would you declare a function template for displaying an array?
6. How do you define a function template when all the parameters are of same type?
7. What is a function template specialization?
8. What is a class template specialization?
9. How do you declare function template which involves more than one type parameters?
10. Can a function template be overloaded? If so how can it be done?
11. How do you define a friend function template for functions having more than one type of
parameters?
12. How do you define a class template for multiple type object data?
13. Give an example of declaration of a friend function template to a class template.
14. How do you declare a class template as friend to another class template?
15. Why is class template called parameterized type?
16. Make a program for a function template for calculation of area of regular polygons. Make a test
program and calculate areas of a square and hexagon having side =10 units.
17. Make a template program to determine greater of two numbers of different types.
18. Make a function template program which can swap two variables which may be int, floats or
characters.
19. Make function template program to swap elements of arrays which may be arrays of integers,
arrays of characters or arrays of double numbers.
20. Make program for display function template for output of integers, floating point numbers,
characters and strings.
21. Make a program for function template to find out and display the smaller of the two numbers
which may be integers or floating decimal point numbers or characters.
22. Make a program for class template by name Cubic which finds out the surface area of
rectangular prismatic components.

❍❍❍
CHAPTER

++

17.1 INTRODUCTION

Standard Template Library (STL) has defined three types of containers, i.e. first class containers,
adapters and near containers (see Chapter 21). There are three near containers which have some
functions similar to other containers and thus some aspects of their behaviour are similar to those
of containers. For instance, these containers support iterators and have functions like begin ()
and end (), etc. The near-containers are bitsets, valarray and C++ strings. We have already discussed
the C-strings in Chapter 10 which are pointer based arrays terminated by Null character. C++
strings belongs to a class template string. Its header file is <string>. C++ strings are functionally
an improvement over C-strings. For example, many of the operators commonly used on
fundamental type variables cannot be used on C-strings, though the individual elements of a C-
string may be treated as elements of an array. Above all, a C-string is a fixed length string. Its
length cannot be changed after its declaration. All these limitations are not there in C++ strings.
The C++ strings can be assigned as a whole and can be modified during execution of program.
The operators =, !=, ==, >, <, >= , <=, +, << and >> can be applied to C++ strings in the same
way as we apply them on variables of fundamental types. A C++ string is a dynamic string, i.e.
the length (number of elements) may be changed during execution of the program. For working
with C++ strings we have to include the header file <string> in the program. A C++ string is
declared as below.
string identifier ;
Here string is the name of class or type of string object and identifier is name of object.

17.2 CONSTRUCTION OF C++ STRINGS

A number of constructors are provided in the string class. C++ string may be constructed by any
of the following methods.
(i) By declaring an empty string.
(ii) By declaring and initializing with string of characters in double quotes.
(iii) By assigning a number of copies of a character to the string.
(iv) As a copy of another string.
(v) As substring of current string by specifying index location and number of characters.
(vi) A substring of current string by specifying the iterator start to iterator end.
❖ 418 ❖ Programming with C++

The methods mentioned above are illustrated in Program 17.1 below. In this program the
string S is not initialized so it is string without any element. This is tested with function
empty(). The string Str1 is declared and initialized by a string of characters in double quotes.
String Str2 is declared to have 6 elements each equal to ‘B’. String Str3 is copy of string Str1.
Str4 is a substring of Str1. It contains 7 characters of string Str1 starting from its 8th character.
String 5 is a substring of Str1 and has elements starting from element at the location
iterator(Str1.begin()+3), i.e. from 4th element to iterator Str1.end(), i.e. to end of Str1.
String Str6 is constructed by the first 8 characters of another string given in double quotes.

PROGRAM 17.1 – Illustrates methods of constructing C++ strings.

#include <iostream>
#include <string> //include header file <string>
using namespace std;
int main(){
string S ;
if(S.empty()) // Test if the string S is empty
cout<< “S is empty”<<endl;
string Str1;
Str1 = “YOU ARE WELCOME!”; /* Initialization by characters in double quotes
(“ “) */
string Str2 (6,’B’); // Str2 is constructed by 6 ‘B’s
string Str3 (Str1); // Str3 is a copy of Str1
string Str4(Str1,8, 7); // Str4 is constructed by 7 characters
//of Str1 starting from 8th character.
string Str5 (Str1.begin()+3, Str1.end()); /*Str5 constructed by iterator
(begin()+ 3) i.e., 4th element of Str1 to end.*/
string Str6 ( “YOU ARE WELCOME!”, 8);
//Str6 is constructed by first 8 characters of string in double quotes
cout<<“Str1 = ”<<Str1<<endl;
cout<<“Str2 = ”<<Str2<<endl;;
cout<<“Str3 = ”<<Str3<<endl;
cout<<“Str4 = ”<<Str4<<endl;
cout<<“Str5 = ”<<Str5<<endl;
cout<<“Str6 = ”<<Str6<<endl;
return 0;
}

The expected output is given below. The output is made self explanatory with the comments
included in the program.
 C++ Strings ❖ 419❖

S is empty
Str1 = YOU ARE WELCOME!
Str2 = BBBBBB
Str3 = YOU ARE WELCOME!
Str4 = WELCOME
Str5 = ARE WELCOME!
Str6 = YOU ARE

17.3 C++ STRING CLASS FUNCTIONS

A number of functions supported by the C++ strings along with brief description of each are
listed in Table 17.1. Below in Program 17.2 we illustrate the application of one of the functions,
i.e. compare() which is used to compare two strings or substrings. The function returns –1,
0 and 1 respectively if the string being compared with the current string is lexicographically
(dictionary style) is greater than, equal to or less than the current string. For example Z is
lexicographically greater than any alphabet between A to Y. Similarly S is greater than K. See the
following program.

PROGRAM 17.2 – Illustrates application of function compare() to strings.

#include <iostream>
#include <string>
using namespace std;

int main()
{string Str1, Str2, Str3, Str4 ;
Str1 = “Calcutta”;
Str2 = “Delhi”;
Str3 = “Delhi”;
Str4 = “ZA”;

cout<< “Str1 = ” << Str1 <<“, Str2 = ” << Str2 <<endl;

cout<<“Str3 = ” << Str3 <<“, Str4 = ” << Str4 <<endl;
cout << “Str1.compare (Str2) = ”<< Str1.compare (Str2)<<endl;

cout << “Str2.compare (Str3) = ” <<Str2.compare (Str3)<<endl;

cout << “Str4.compare (Str2) = ” <<Str4.compare (Str2)<<endl;
cout << “Str3.compare (Str4) = ” <<Str3.compare (Str4)<<endl;
return 0;
}
❖ 420 ❖ Programming with C++

The expected output is given below.

Str1 = Calcutta, Str2 = Delhi
Str3 = Delhi, Str4 = ZA
Str1.compare (Str2) = –1
Str2.compare (Str3) = 0
Str4.compare (Str2) = 1
Str3.compare (Str4) = –1
Among the above three strings, ZA is greatest of all, Z is greater than all other alphabets in
upper case. Similarly Delhi is greater than Calcutta, because D is greater than C. So the result is
quite obvious.
Table 17.1 – C++ string class functions and their description

Function Description
append() The function is used to append a substring at the end of current string. For
details see Program17.3.
assign() The function assigns (i) a string or (ii) a specified number of characters of a
string or (iii) a substring to the current string. For details see Program 17.3.
at() The function returns reference to an element at the indexed location. See
Program 17.3 for illustration.
begin() Returns iterator to the first element of the string. See Programs 17.1& 17.4
C_str() Returns a const pointer to a C-string which is identical to the current C++
string. C-strings are terminated by Null character.
capacity() Returns the maximum number of elements that the current memory
allocation can hold. See Program 17.3 for its application.
clear() The function deletes all elements of the string.
compare() The function returns –1, 0 and 1 respectively if the string being compared
lexicographically (i.e. dictionary style) is bigger than, equal to or smaller than
the current string. For illustration see Program 17.2 and 17.4.
copy() The function copies specified number of characters of current string. For
illustration see Program 17.6.
data() Function returns a pointer to the first element of current string. See Program
17.6 for illustration.
empty() Returns true if there are no elements in the string. Returns false if string is
not empty. See Program 17.1 and Program 17.4 for illustration.
end() It returns iterator pointing to just past the end of string. See Program 17.1
and Program 17.4 for illustration.
erase() The function erases specified elements of the string. For illustration see
Program 17.4.
Contd...
 C++ Strings ❖ 421❖

Function Description
find() Returns location of first occurrence of a string/substring in the
current string. For illustration see Program 17.6.
find_first_not_of() It returns index of the first occurrence of a character of current
string which does not match any character of specified string.
find_first_of () The function returns index of first occurrence of a character that
matches a character of given string.
find_last_not_of() The function returns index of the last occurrence of a character
of current string which does not match any character of given
string.
find_last_of() The function returns index of last occurrence of a character that
matches a character of given string.
getline() The function is used to read strings from I/O stream.
insert() Function is used to insert characters or substring in various ways
in a current string.
length() It returns the number of elements in the string without Null
character.
max_size() Returns the maximum number of elements that can be held in
the presently allocated memory for the string.
rbegin() It returns a reverse iterator to end of string.
rend() It returns a reverse iterator to beginning of string.
replace() Function is used to replace elements in various ways.
reserve() It sets the capacity of string to minimum size.
resize() Function changes the size of string to specified size. If the value
of elements is also specified the newly created elements are
initialized to that value.
rfind() Function returns the first occurrence of string in the current
string by reverse search from given index.
size() Function returns current number of elements in the string. See
Program 17.3 and Program 17.4 for application.
substr() It returns substring of specified number of elements (n) out of
the current string from the specified location. ( Program 17.4).
swap() It exchanges (swaps) the elements of two strings.

17.4 APPLICATIONS OF SOME C++ STRING CLASS FUNCTIONS

The application of functions append(), assign (), at(), capacity()and size()is

illustrated in the following program. The function size() gives the actual number of elements in
the string while capacity gives the maximum number of elements that may be held in the current
allocation of memory for the string.
❖ 422 ❖ Programming with C++
PROGRAM 17.3 – Illustrates application of functions append(), assign (), at(),
capacity() and size ().

#include <iostream>
#include <string>
using namespace std;
int main()
{string Str1, Str2, Str3 ;
Str1 = “you are welcome!”;
Str2 = “If you want to learn C++,”;
Str3 = “If you want to join this college,”;

cout<< “Size of Str1 = ”<<Str1.size ()<<endl;

cout<< “Capacity of Str1 = ” << Str1.capacity()<<endl;
cout<<“Before appending \nStr1 =”<< Str1;
cout<<“\nStr2 = ” << Str2<<“\nStr3 = ”<< Str3<<endl;

Str2.append(Str1); // Str1 is appended at back of Str2

Str3.append (Str1); // Str1 is appended at back of Str3

cout<<“After appending\nStr2 = ”<< Str2 <<“\n”<<“Str3 = “<< Str3 <<endl;

Str1.assign (Str2); // Str2 is assigned to Str1

cout<< “After assigning \nStr1 = ”<<Str1<<endl;
cout<<“The element at location 5 of Str2 is ”<<Str2.at(5)<< endl;
return 0 ;
}

The expected output is as under.

Size of Str1 = 16
Capacity of Str1 = 31
Before appending
Str1 = you are welcome!
Str2 = If you want to learn C++,
Str3 = If you want to join this college,
After appending

Str2 = If you want to learn C++, you are welcome!

Str3 = If you want to join this college, you are welcome!
After assigning
Str1 = If you want to learn C++, you are welcome!
The element at location 5 of Str2 is u
 C++ Strings ❖ 423❖
The output given above is self explanatory. The output shows the difference between
size() and capacity(). The statement Str1.size() gives the current number of
elements in Str1 as 16 while it has capacity for 31 elements.

FUNCTION ERASE()

The function may be used for the following.

(i) To erase the entire string the code is given below and an illustration is given in Program
17.4.
Str.erase();
(ii) To erase a specific number of characters starting from a specified location. Say we want
to erase 3 characters of string Str starting from index 4. The code is as below.
Str.erase(4,3);
(iii) To erase all character of string Str starting from a position (say 10) to end of string. The
code is given below.
Str.erase (10);

PROGRAM 17.4 – Illustrates substr(), begin(),end(),compare()and erase().

#include <iostream>
#include <string>
using namespace std;
int main()
{
string Str1, Str2, Str3 ;
Str1 = “You are welcome to this meeting”;
Str3 = “Mahrashtra”;
cout << “Size of Str1 = ”<<Str1.size ();
cout << “\nSize of Str3 = ”<<Str3.size();
cout<<“\nInitial Str1 = “<<Str1<<endl;
cout<<“Str1.substr(4,11 ) = “<<Str1.substr(4,11)<<endl;
Str2 = “You are going to learn C++” ;
cout<<“Str2 = “<<Str2<<endl;
string :: iterator itr1 =Str2.begin() ; // for iterators see
string :: iterator itr2 = Str2.end(); //Chapter21, Section 21.3
Str2.erase (itr1+3, itr1+16);
cout<<“After Str2.erase ( itr1+3, itr1+16) = “<<Str2<<endl;
Str2.erase (4);
cout<<“After Str2.erase(4) = “ <<Str2<<endl;
Str2.erase();
❖ 424 ❖ Programming with C++
if ( Str2.empty ())
cout<<“After Str2.erase() Str2 is empty.”<<endl;
return 0;
}

The expected output is given below.

Size of Str1 = 31
Size of Str3 =10
Initial Str1 = You are welcome to this meeting
Str1.substr(4,11 ) = are welcome
Str2 = You are going to learn C++
After Str2.erase (itr1+3, itr1+16) = You learn C++
After Str2.erase(4) = You
After Str2.erase() Str2 is empty.

The first two lines of output give the size of string Str1 and string Str3. The fourth line of
output consists of a substring of Str1 which contains 11 characters starting from 4th character.
In the next line of output the effect of erasing from 4th element to 17th element of Str2 leaves
You learn C++. It is erased again from 4th element to end, so only You is left. It is again erased
by erase() which erases it completely.

17.5 STRING CLASS OPERATORS

The following operators are supported by class string.
= , ==, !=, < , > , <=, >= , + , << and >>
Let us have three strings S1 , S2 and S3 as declared below.
string S1 (“welcome!”) ;
string S2 (“You are”) , S3 ;
The operator + may be used as it is used with fundamental types ( See Prog.17.5). Thus we can
write
S3 = S2 + S1;
The above operator + appends S1 at back of S2 and we get the following.
S3 = “You are welcome!”;
Similarly we may add a character to a string. Thus, let string S be given as
string S = orange;
The operation S +‘s’ will add s at end of string S. After the addition S = “oranges”.

FUNCTION GETLINE()

The function getline() for a string is coded as below,

string Str;
getline (cin, str) ;
The application of the function is illustrated in the following program. The program also
illustrates application of some operators on strings.
 C++ Strings ❖ 425❖
PROGRAM 17.5 – Illustrates action of operators on strings.

#include <iostream>
#include <string>
using namespace std;
int main()
{
string Str1(“ YOU ARE WELCOME!”), Str2(“Nikita,”) ;
cout << “Str1 = ”<<Str1<<endl;
string Str3;
Str3= Str2+Str1;
cout<<“Str3 = ”<< Str3 << endl;
string S1 = “Orange”;
string S2 = “ointer”;
string S3 = “P”;
char ch = ‘s’;
string S4= S1 + ch;
S2 = S3+S2;
cout<<“S1 = ”<< S1<<“\tS4 =”<<S4 <<“\tS2 = ”<<S2 <<endl;
string S5;
cout<<“Write a small sentence\n”;
getline(cin, S5);
cout<<“S5 = ”<<S5<<endl;
if ( Str1> S5)
cout<< “Str1 is greater than S5”<<endl;
else
cout <<“Str5 is bigger than S1.”<<endl;
if (S1==Str1)
cout<< “S1 and Str1 are equal”<<endl;
else
cout<< “S1 and Str1 are not equal”<<endl;
if (S4>S1)
cout << “S4 is greater than S1”<<endl;
else
cout << “S1 is greater than S4”<<endl;
return 0;
}

The expected output is given below. The output is self explanatory.

Str1 = YOU ARE WELCOME!
Str3 = Nikita, YOU ARE WELCOME!
S1 = Orange S4 =Oranges S2 = Pointer
❖ 426 ❖ Programming with C++
Write a small sentence
He is going to office.

S5 = He is going to office.
Str5 is bigger than S1.
S1 and Str1 are not equal
S4 is greater than S1

FUNCTIONS DATA(), FIND(), COPY()

The code Str.data() returns pointer to the first element of the string Str. Function find()
returns the position of first occurrence of a string/substring in the current string. The function
copy() makes a copy of specified number of characters of a string. The following program
illustrates the application of the two functions along the function compare used in if condition.

PROGRAM 17.6 – Illustrates the application of functions data(), find() and copy().

#include <iostream>
#include <string>
using namespace std;
int main()
{
string Str1, Str2, Str3, Str4 ;

Str1 = “You are welcome!”;

Str2 = Str1; // assignment
cout<<“Str2 = ”<<Str2;

Str3 = “If you want to join this college,”;

cout<<Str3;
cout<< “\ndata = ”<< Str2.data();

if ( (Str1.compare(Str2)) ==0)
cout << “\nStr1 and Str2 are equal”<<endl;
else
cout << “Str1 and Str2 are not equal” <<endl;
cout<<“In Str1 substring com occurs at ”<<Str1.find(“com”)<<endl;
char S[12]; // C-type string or array of characters
memset(S,’\0',12) ; //The function memset()fills the 12
//elements of array S with Null character

Str3.copy( S, 11); //copy first 11 characters of Str3

 C++ Strings ❖ 427❖
cout<<“Number of characters copied = ”<< Str3.copy( S ,11)<<endl;
// Display number of characters copied
cout<<S<<endl; // Display the C-string S
cout<<Str3<<endl; // Display Str3
return 0;
}

The expected output is given below.

Str2 = You are welcome!
If you want to join this college,
data = You are welcome!
Str1 and Str2 are equal
In Str1 substring com occurs at 11
Number of characters copied = 11
If you wan
If you want to join this college,

FUNCTION find_first_not_of( ) , find_last_not_of( )

find_first_of( ), find_last_of() and replace()

The code for the first function for Str1 and Str2 would be coded as
Str2.find_first_not_of(Str1);

The return value of the function is the index value of the first element of Str2 which is not
an element of Str1. The second function may be coded as
Str2.find_last_not_of(Str1);
This function returns the index value of the last element of Str2 which is not a member of
Str1. The application of these and the other two functions mentioned above are illustrated in
the following program.

PROGRAM 17.7 – Illustrates the application of functions find_first_not_of(),

find_last_ not_of().
#include <iostream>
#include <string>
using namespace std;
int main()
{
string Str1, Str2, Str3(6,’B’) ;
Str1 = “YOU ARE WELCOME!”;
Str2 = “Are you going to DELHI”;

cout<<( Str2.find_first_not_of(Str1))<<endl;
cout<<( Str2.find_last_not_of(Str1))<<endl;
❖ 428 ❖ Programming with C++
cout<<( Str2.find_first_of(Str1))<<endl;
cout<<( Str2.find_last_of(Str1))<<endl;
Str3.replace( 1,5 ,“B.L.Juneja”);
//replace 5 elements Str3 starting from index 1.
cout<< “Now Str3 = ”<< Str3<<endl;
return 0; }

The expected output is as under.

1 //Comments: the first element is r with index value 1
21 // The last element is I with index value 21.
0 // The first element is A with index value 0
19 // The last element is L with index value 19.
Now Str3 = B.L.Juneja //The 0th element is B other Bs are replaced

17.6 ARRAY OF C++ STRINGS

An array of strings may be declared as

string S[n];
Here string is the type (name of class), S is the name of array and n is the number of strings
in the array. Each element of the array is accessed by its index value in the array. The application
is illustrated in the following program.
PROGRAM 17.8 – Illustrates array of strings.

#include <iostream>
#include <string>
using namespace std;
int main()
{ string S1, S2[4] ,S3;
string Name[4] = { “Nimi”, “Malti”, “Narayani”, “Shanti”};
for (int i =0; i < 4; i++)
S2[i] = Name[i]; // assigning array elements to S2

for( int k=0; k<4; k++)

cout<<S2[k]<<“ ”; // output of array elements of S2
S1 = “ , You are welcome to C++! ”;
string Msg [4];
cout<<“\n”;

for( int j =0; j<4; j++)

{Msg[j] = S2[j]. append (S1); // append S1 to elements of S2
// and assign the result to elements of Msg[]
cout << “Msg [“<<j<<”] = ” <<Msg[j]<<endl; }

return 0;
}
 C++ Strings ❖ 429❖
The expected output is given below.
Nimi Malti Narayani Shanti
Msg [0] = Nimi , You are welcome to C++!
Msg [1] = Malti , You are welcome to C++!
Msg [2] = Narayani , You are welcome to C++!
Msg [3] = Shanti , You are welcome to C++!

The output is self explanatory. The program takes an array of names and appends a message
at the end of each element (name) of string array. The program also illustrates the assignment
of strings.

EXERCISES
1. What is the difference between C-string and C++ string?
2. Do the C++ strings have null character at the end?
3. What are the methods of constructing C++ strings?
4. Which operators can be applied to C++ strings?
5. Do the C-strings also support the operators that are supported by C++ strings?
6. What is the difference between function sizeof() and function capacity ()?
7. What action the following C++ string functions perform?
(i) append () (ii) assign ()
(iii) capacity () (iv) size ()
8. What actions the following C++ string functions perform?
(i) at () (ii) compare ()
(iii) copy ()
9. What do the following functions of C++ string do?
(i) data () (ii) find ()
(iii) find_first_of () (iv) find_last_of
10. Make program to illustrate the action of following functions of class string.
(i) reserve () (ii) replace ()
(iii) max_size()
11. Make a program to illustrate following functions for C++ strings.
(i) getline () (ii) insert ()
(iii) length ()
12. Make a program to compare the action of function length() of C++ string and function
sizeof() of C-string.
13. How do you convert a C++ string into a C-string?
14. Make program to illustrate the application of following functions of class string.
(i) clear () (ii) resize ()
(iii) refind ()
❖ 430 ❖ Programming with C++
15. What do you understand by the following statement involving the function erase(). Str is the
name of string.
(i) Str.erase (4,3); (ii) Str.erase (12);
(iii) Str.erase ()
16. What action takes place when code str. append (Str2); is executed.
17. Make a program to illustrate the functions length() and size().
18. What is the difference between the string class function size() and the C++ Standard Library
function sizeof()?
19. Make a program to illustrate the action of operators ==, != and = on strings.

Answer: The following program illustrates the operators.

PROGRAM 17.9: Illustrates use of relational operators on strings.

#include <iostream>
# include<string>
using namespace std;
int main()
{
string Str1 = “Hello! Good Morning” ;
string Str2 = “Same to you”;
string Str3 = “Let us learn C++”;
if( Str3 !=Str2)
cout<<“Str3 and Str2 are different.”<<endl;
cout << ( Str1 ==Str2 ? “True “: “False” )<<“\n”;
cout << “ Before assignment \n Str1 = ”<<Str1<<endl;
cout << “ and \n Str2 = ”<<Str2 <<endl;
Str3 = Str2 = Str1;
cout << “After the assignment”<<endl;
cout<< “ Str1= ”<<Str1<<“\n Str2 = ”<<Str2 <<“\n Str3 = ” <<Str3 <<endl;
return 0 ;
}

The output is given below.

Str3 and Str2 are different.
False
Before assignment
Str1 = Hello! Good Morning
and
Str2 = Same to you
After the assignment
Str1= Hello! Good Morning
Str2 = Hello! Good Morning
Str3 = Hello! Good Morning

❍❍❍
CHAPTER

18.1 INTRODUCTION

The present day society is increasing becoming dependant on computer systems. Whether it is
reservation and ticketing for travel by air, train or bus, telephonic communication by land lines
and mobiles, transactions in banks, transactions in stock exchanges, keeping of records by
companies as well as government agencies, raising and collection of bills, the systems of credit
cards and ATMs, internet system, admission of students in universities, etc., are all controlled
by computer systems. So much so that even in sports the decisions of winner or loser in split
second finishes are decided by replaying the computer records made during the event. In such
an environment it is absolutely necessary that the computer systems (hardware and software)
handling these tasks should be highly reliable and should function as expected continuously.
However, exceptions do happen either due to failure of a component of hardware system or due
to failure of a portion of software/program or due to some unexpected circumstance created
during running of the program. The exceptions created due to causes which are unrelated to the
program such as hardware problems, usually called asynchronous exceptions are not in the scope
of present chapter. In this chapter we shall deal with synchronous exceptions which are caused in
a program which otherwise runs normally, but due to some exceptional input or exceptional
conditions like non-availability of computer memory, division by zero etc., runs into problems
which may lead to aborting the program.
In C++ language, there is provision to deal with many of the known types of exceptions. A
programmer may add exception handling code at the places in the program, where it is feared
that problems may arise. If an exception occurs the exception handling code takes care of the
problem. Normally the exception handling provision informs the user about the exception (error)
when it occurs and also prompts the user for a proper corrective action. For instance, if an
improper input is the cause of problem it will ask the user to correct the input data. For other
problems remedial measures are provided in the exception handling code included in the
program. For some types of errors, the program may also be made fault tolerant. In such cases
the program may run normally though in a part, faults have occurred. However, if the exception
cannot be handled the exception handling code would call the function abort() to terminate the
program. In this process, however, there are functions which may be used to inform the user
about the type of error that has occurred and that the program is being aborted.
❖ 432 ❖ Programming with C++
Aborting a program is an extreme step which normally should not be resorted to if it can
be managed otherwise. You can think of the problems if the software of a bank fails. In C++
there are provisions to deal with the known exceptions. Below we discuss the types of exceptions
and how to deal with them.

18.2 THE try, throw and catch

In C++ the exception handling is managed through three keywords, i.e. try, throw and catch.
The programmer has to identify the portions of program where an exception is most likely to
occur. For example, user input portion can be one such region where it is likely that a user may
enter wrong data. Provision is made so that the user can correct the input data. Another example
is that in an integer division in which the denominator may become zero or in floating point
numbers the denominator may become a very small quantity leading to overflow problem. The
statements for such a portion are placed in a try block. A try block starts with keyword try which
is followed by the statements which are placed in curly braces {}. The try block, generally
comprises selection conditions for the values of variable that may lead to an exception. That is
the detection mechanism. If an exception is detected it is thrown by using the keyword throw
as illustrated in the sample code below.
try
{statements; // try and throw block
throw type_exception; //if exception is detected, throw the type
}
catch (type1_exception) // first catch block
{
statements;
}
catch(type2_exception) // second catch block
{
statements;
}
Note that the throw segment is immediately followed by one or more catch blocks. Each
catch block starts with the keyword catch followed by parentheses () which contains the type
of the catch block, this is followed by statements of the block which are enclosed between curly
braces {}. Each catch block has a different type to catch. Here type refers to either one of the
fundamental data types such as int, float, double or char or the user defined types, i.e. the name
of class of the object thrown. The selection of a particular catch block depends on the result of
matching the type of the throw with the type of catch block. Obviously there should not be two
catch blocks with same type. The whole process is illustrated in the Figure 18.1.
 Exception Handling ❖ 433❖

try { statements;

throw (char); }

catch blocks
catch (int)
Match type of throw
with type of catch {exception handling
mechanism1

catch (char)

{exception handling
mechanism2

Fig. 18.1: Illustration of try, throw and catch mechanism

The following program illustrates the exception that may occur if division by zero is tried.

PROGRAM 18.1 – Illustrates application of try, throw and catch.

#include<iostream>
using namespace std;
int main ()
{ int A, B ;
double D;
cout<<“Enter two numbers ”;
cin>> A>>B;

try // try block

{ if( B == 0)
throw B; // throw
else
{ D = A/double(B);
cout << “D = ” << D <<endl; }} // End of try block

catch (int C ) // exception handling block starts

{ cout << “You have entered B = 0, enter another number”<<endl;
}
return 0;
}

The expected output when B is put as 0 is given below.

Enter two numbers 5 0
You have entered B = 0, enter another number
Second trial with the same program in which B = 15 is entered is given below.
Enter two numbers 6 15
D = 0.4
❖ 434 ❖ Programming with C++
It should be noted that int C is a local variable of the catch block mentioned above. The
catch process is type sensitive. It will only catch the type of throw which matches with the type
declared in catch. The try block may be followed by a number of catch blocks each with a
different type to catch. If the type of throw and that of catch are different, the catch block will
be skipped. In fact, putting in the name of variable such as C in Program 18.1, is not necessary.
We may as well write the statement as catch(int). That is the type of the variable thrown.
The following program is another example which illustrates that if type of object thrown is double
the catch with argument double handles it.

PROGRAM 18.2 – Illustrates that type of throw and type of catch block should match.

#include<iostream>
using namespace std;
int main ()
{ int A ;
double B , D ;
cout<<“Enter two numbers ”;
cin>> A>>B;
try // try block
{ if(B == 0.0)
throw B; // throw statement B is double
else
{ D = A/B;
cout << “D = ” << D <<endl; }} // End of try block
catch (int) // exception handling block for “int” throw
{ cout << “The throw \”int \” is caught”<<endl; }

catch( double) // exception handling block for “double” throw

{ cout <<“The throw \”double\” is caught.”<<endl;}
return 0;
}

The expected output is shown below. The output shows that the catch type which matches with
the type of throw will handle the exception.
Enter two numbers 6 0
The throw “double” is caught.

PROGRAM 18.3 – Illustrates that if none of the catch blocks’ types provided in the
program matches the type of throw, the program will be aborted.

#include<iostream>
using namespace std;
int main ()
{
int A ;
double B , D ;
 Exception Handling ❖ 435❖
char ch = ‘C’;
cout<<“Enter two numbers ”;
cin>> A>>B;
try // try block
{
if(B == 0.0)
throw ch; // throws a char type
else
{ D = A/B ;
cout << “D = ” << D <<endl; }} // End of try block
catch (int) // exception handling block for int
{ cout << “The throw \”int \” is caught”<<endl; }
catch( double) // exception handling block for double
{ cout <<“The throw \”double\” is caught.”<<endl;}
return 0;
}

None of the catch types matches the type of throw, i.e. char. So the program is aborted. The
following is shown on monitor.
abnormal program termination.
In the above program, the type of throw is char while the catch types provided are int and
double. So neither of the two catch types match thus none of them would work. In such a case
the function abort () is called to close the program. The program also shows that the type of throw
need not be the type of variable involved in the exception. The following program shows that
the exception thrown may be a class object. In that case the name of class is the type of exception
thrown. When a class object is thrown its type is the name of class to which it belongs.

PROGRAM 18.4 – Illustrates that throw is a class object.

#include<iostream>
using namespace std;
class Base // class Base defined
{ };
int main ()
{ Base base_obj ; //base_obj is an Object of class Base
int A ;
double B , D ;
char ch = ‘C’;
cout<<“Enter two numbers ”;
cin>> A>>B;
❖ 436 ❖ Programming with C++
try // try block
{
if(B == 0.0)
throw base_obj; // throws an object of class Base
else
{ D = A/B ;
cout << “D = “ << D <<endl; }} // End of try block
catch (int) // exception handling block for int
{ cout << “The throw \”int \” is caught”<<endl; }
catch(Base) // exception handling block for class Base
{ cout <<“The throw \”Base\” is caught”<<endl;}
return 0;
}

The expected output is given below.

Enter two numbers 6 0
The throw “Base” is caught
If the exception thrown is a derived class object, the catch block with base class type can
handle it. It is illustrated in Program 18.17 in the exercise 15.
In the following program a class with name Except is defined. It is different from the C++
Std. Library class except. The throw is used with an object Exc of class Except. In the catch
statement the type is same, i.e. Except, however the object now is Exd. The exception is caught
because type is same. If more than one exceptions of the same type are thrown, then only one
catch block is needed for them. In order to differentiate the two exceptions the selection
statements are provided in the catch block so that appropriate exception handler may be invoked.

PROGRAM 18.5 – Illustrates that multiple exceptions may be thrown from same try block.

# include<iostream>
using namespace std;
class Except {
public:
void Display1( )
{cout<<“There is an exception.\nValue entered for A is negative\n ”;
cout <<“Enter a positive value.\n”; }
void Display2 ()
{cout<<“The number B entered is zero. \n”;
cout << “Enter a number more than zero.\n” ;}
};
int main ()
{ Except Exc;
 Exception Handling ❖ 437❖
int A, B ;
double D;
cout<<“Enter two numbers ”;
cin>> A>>B;
try
{
if (B == 0)
throw Exc ;
else
if ( A < 0 )
throw Exc;
else
{D = A/ double(B);
cout << “D = ” << D <<endl;
}}
catch (Except Exd )
{if ( A < 0)
Exd.Display1();
if ( B == 0 )
Exd.Display2();
}
return 0;
}

Two trial runs are carried out. In the first case both the exceptions are thrown and in the
second case none of the exceptions is thrown. The expected outputs in the two cases are shown
below. Output of first trial is given below.
Enter two numbers 4 0
There is an exception.
Value entered for A is negative
Enter a positive value.
The number B entered is zero.
Enter a number more than zero.
Output of second trial is given below.
Enter two numbers 10 5.0
D=2

The try block may have more than one throw statements. Also try block may be followed
by more than one catch blocks but their arguments should of different type. One catch block
can not have two different types as arguments, for example the code like catch (int A,
double B) will result in compiler error. In the following program two different types are
thrown in the same try block, but it needs two separate catch blocks, one for each type.
❖ 438 ❖ Programming with C++
PROGRAM 18.6 – Illustrates more than one throw from same try block, needing different
catch blocks.

#include<iostream>
using namespace std;
int main ()
{ int A ;
double D, B;
cout<<“Enter two numbers ”;
cin>> A>>B;
try
{
if (B == 0)
throw B;
if (A > 50)
throw A;
else
{D = A/ B;
cout << “D = ” << D <<endl;
}}
catch (double C )
{cout << “You have entered B = 0, enter another number”<<endl;}
catch (int E )
{cout<< “Enter another number less than 50 for A. ”<<endl; }
return 0;
}

The expected output is given below.

Enter two numbers 60 5
Enter another number less than 50 for A.

In the same try block, one may have multiple throws. If the throws are of same type only
one catch block is needed. For the information of the user as to which exception is caught, a set
of selection statements may be included in the catch block. As is normal with selection
statements if one statement is selected the others are neglected.

18.3 CATCH ALL TYPES OF EXCEPTIONS

For catching all types of exceptions thrown, the code catch (…) may be used. Note that there
are only three points in the parentheses. With more or less number of these the compiler may
show error. The following program illustrates catch (…) .
 Exception Handling ❖ 439❖
PROGRAM 18.7 – Illustrates catch(…), i.e. catch all types of exceptions.

#include<iostream>
using namespace std;
class Base
{ };
int main ()
{ Base base_obj ;
int A ;
double B , D ;
char ch ;
cout<<“Enter an integer, a double and a character: ”;
cin>> A>>B>>ch;
try
{ if(B == 0.0)
throw base_obj;
if (A > 50)
throw A;
if(ch != ‘Z’)
throw ch;
else
{ D = A/double(B);
cout << “D = ” << D <<endl; }} // End of try block
catch (...) // exception handling block starts
{
if ( A >50)
cout << “The exception on int is caught”<<endl;
if (B==0)
cout <<“The exception on Base is caught.”<<endl;

if( ch != ‘Z’)
cout <<“The exception on char is caught.”<<endl; }
return 0;
}

The program has been run twice to test it. In the first trial all the variables are in the range so
the program runs normally and following is the output.
Enter an integer, a double and a character: 40 6.0 Z
D = 6.66667
In second trial all the variables are out of range and following output is obtained.
❖ 440 ❖ Programming with C++
Enter an integer, a double and a character: 56 0 H
The exception on int is caught
The exception on Base is caught.
The exception on char is caught.

18.4 EXCEPTION HANDLING FUNCTION

The try-throw and catch sequence may be put in a function. It is convenient to call the function
in the main program. The following program illustrates the same.

PROGRAM 18.8 – Illustrates application of exception handling function.

#include<iostream>
using namespace std;
void Test(int A ,double B, char ch)
{ try

{
if ( A > 50.0)
{cout << “A not in range”<<endl;
throw A; }
if (B == 0)
{cout <<“B is equal to zero”<<endl;
throw B; }
if (ch != ‘D’)
{cout<< “ch is not equal to D”<<endl;
throw ch; }
else
cout<< “All in range. No exception thrown:” <<endl; }

catch (...)
{ cout<< “Caught an exception ”<<endl; }
}

int main ()
{ int A;
double B;
char ch;
cout<< “Write values of A, B and ch :”;
cin >> A>>B>>ch;
Test (A,B,ch); // Test function for try, throw and catch
return 0;
}

The above program has been run 4 times in order to test its operation for different types
of exceptions. The first trial is for a case when all variable are in range and no exception is thrown.
The output of this is given below.
 Exception Handling ❖ 441❖
Write values of A, B and ch :40 6.0 D
All in range. No exception thrown :
The second case relates to when only A is not in range, so exception thrown is of type integer.
The output is as under.
Write values of A, B and ch :60 5.0 D
A not in range
Caught an exception
In the third case only B is equal to 0 while other two are in range. So exception thrown is
double. The output is as under.
Write values of A, B and ch :40 0.0 D
B is equal to zero
Caught an exception
In the fourth case the error is in value of character. So an exception of type character is thrown
and caught as illustrated in the following output.
Write values of A, B and ch :40 5.0 C
ch is not equal to D
Caught an exception

18.5 EXCEPTION SPECIFICATION

C++ allows the user to declare a set of exceptions that a function may throw. The exception
specification is in fact is a suffix to the normal function head. The suffix comprises the keyword
throw followed by, in parentheses, the list of exception types. A function definition with an
exception specification is of the following general form.
type function_identifier ( argument_List) throw ( exception_type_list ) // function head
( statements;} // function body
In the above definition the first word type is the type of data returned by the function. This
is followed by the function name which is followed by arguments list in parentheses and then
the suffix which comprises the keyword throw followed by a list of exception types in parentheses.
The function type is not affected by the suffix. The exception_type_list as well as the argument_
list are optional and hence we may have the following three cases.
(i) void Function1() throw () // Empty exception_type_list, cannot throw an exception.
(ii) void Function2 () throw ( X, Y, Z) // Can throw X, Y and Z types of exceptions only.
(iii) void Function3 () // Can throw any type of exception.
If the function throws an exception which is not in the exception_type_list the function
unexpected() is called. The default action of the function unexpected is to call the function
abort() which terminates the program.
The function with suffix throw() cannot throw any exception and if it does the function
unexpected is called and the program is aborted. Similarly if the second function with
exception_type_list ( X, Y , Z) throws an exception other than X, Y or Z the program would
❖ 442 ❖ Programming with C++
be terminated. In the third option ( void Function3()), the function can throw any type of
exception.

PROGRAM 18.9 – Illustrates exception specification.

#include<iostream>
using namespace std;
void Fexcep (int j ) throw () // cannot throw an exception
{
if ( j== 1)
throw j;

if ( j == 2)
throw char () ;
if ( j==3)
throw double();}

int main ()
{
int n;
cout << “Enter a number from 1 to 3: “;
cin >> n ;
try{
Fexcep(n);
} // End of try block

catch (int ) // exception handling block starts

{ cout << “The throw int is caught”<<endl;
}
catch(double)
{ cout <<“The throw double is caught.”<<endl;}

return 0;
}

On execution a number 2 is entered, and this throws an exception of type char. But the function
is with empty exception list. Therefore, the program is terminated. The output is given below.
Enter a number from 1 to 3: 2
abnormal program termination

PROGRAM 18.10 – Program illustrates that if an exception of the type thrown is not in the
exception specification list of the function, the program is terminated.

#include<iostream>
using namespace std;
 Exception Handling ❖ 443❖
//The following function can throw int or double
void Fexcep (int j ) throw (int, double)
{
if (j== 1)
throw j;

if (j == 2)
throw ‘D’ ;

if(j==3)
throw 0.8;}

int main ()
{ int n;

cout << “Enter a number from 1 to 3: ”;

cin >> n ;
try
{
Fexcep(n);
} // End of try block

catch (int ) // exception handling block starts

{ cout << “The throw int is caught”<<endl;
}
catch(double)
{ cout <<“The throw double is caught.”<<endl;}

return 0;
}

The program is run and number 2 is entered. This corresponds to throwing a char which is not
in the specification list. Therefore the program is terminated. The output is given below.
Enter a number from 1 to 3: 2
On entering 2, the program is terminated with following message.
abnormal program termination

In the following the exception thrown are implemented because they are in the specification
list.

PROGRAM 18.11 – Illustrates implementation of specified exceptions.

#include<iostream>
using namespace std;
void Functspecify ( int j ) throw ( int, double, char)
{
if ( j == 1)
throw j;
❖ 444 ❖ Programming with C++

if ( j ==2)
throw 0.6;
if ( j == 3)
throw ‘H’;}
int main ()
{
try
{
int A (4);
if ( A != 3)
Functspecify (1);}
catch (int ) // exception handling block starts
{ cout << “The exception on int is caught”<<endl;
}
try { // try block
double B (6.6) ;
if ( B!=7)
Functspecify ( 2 );}
catch (double ) // exception handling block starts
{ cout << “The exception on double is caught”<<endl;
}
try {
char ch = ‘C’;
if ( ch != ‘H’)
Functspecify ( 3 ) ;}
catch(char)
{ cout <<“The exception on char is caught.”<<endl;}
return 0;
}

The expected output is given below.

The exception on int is caught
The exception on double is caught
The exception on char is caught.

18.6 RETHROW AN EXCEPTION

An exception may be caught by a handler and partly dealt with. The programmer may like to
rethrow the exception so that it can be dealt again by an outer catch mechanism. For such a case
the following code may be included in the catch block, i.e. throw without parentheses.
throw ;
 Exception Handling ❖ 445❖
This code will rethrow the same type of exception so that it can be dealt by outer catch
block. The rethrown exception is not caught by the same catch block. It is caught by an outer
catch block.
The following program illustrates the rethrow done twice.

PROGRAM 18.12 – Illustrates re-throw of exception.

#include<iostream>
using namespace std;

void Rethrow(int A ) //definition of Function Rethrow()

{
try
{ int B;
cout<<“Enter a positive number : “; cin>> B ;
if ( B<=0)
throw B;}

catch (int)
{ cout<<“This catch is inside function.”<<endl;
throw; // first rethrow
}}

int main ()
{
try{

try {
Rethrow (6);
}

catch(int)
{ cout << “This catch is inner catch inside main.”<<endl;
throw; // second rethrow
}}

catch(int)
{
cout << “This catch is the outer catch in the main.”<<endl;
}
return 0;
}

The expected output is given below.

Enter a positive number : –10
This catch is inside function.
This catch is inner catch inside main.
This catch is the outer catch in the main.
❖ 446 ❖ Programming with C++

18.7 C++ STANDARD LIBRARY EXCEPTION CLASSES

There are a large number of classes in C++ Standard Library which deal with exceptions.
Figure 8.2 shows the hierarchy of prominent exception handling classes in C++. The header file
<exception> defines the base class exception. A number of classes are derived from the base class.
These include classes on logic_error , runtime_error , bad_alloc, bad_cast, bad_typeid and bad
exception. The classes derived from runtime_error are the overflow and underflow errors. Three
classes are derived from logic_error class. These are out_of_range, length_error and
invalid_argument. The classes on runtime error and logic error are defined in header file
<stdexcept>. The four classes, i.e. bad_alloc, bad_cast, bad_typeid and bad_exception are the ones
in which exceptions are thrown by operators. The class except also defines a virtual function what()
which is overridden by every derived class to convey an appropriate message.
The different classes dealing with exceptions and their inherited classes are illustrated in
Fig. 18.2 below. Table 18.1 elaborates the various exception classes.

overflow_error
runtime_error
underflow_error

thrown by
bad_alloc
new operator

thrown by
bad_cast
dynamic_cast
class
exception thrown by
bad_typeid
typeid ( )

thrown by function
bad_exception
unexpected ( )

invalid_argument

logic_error
length_error

out_of_range

Fig.18.2: Exception classes of C++ Standard Library

 Exception Handling ❖ 447❖
Table 18.1

Class Description
overflow_error Deals with overflow of a number if it is too big for the computer.
underflow_error Deals with underflow of number. Number is too small for the computer to store.
bad_alloc Exception thrown by operator new in case of bad allocation of memory ( memory
not available).
bad_cast This exception is thrown by dynamic_cast if it is unsuccessful.
bad_typeid Exception thrown by operator typeid() if it is unsuccessful.
bad_exception If an exception, not listed in exception specification occurs the exception
bad_exception is thrown by function unexpected(). This may be managed by
registering function set_unexpected ().
invalid_argument For example a negative number for a square root is invalid argument.
length_error It indicates that a length larger than the one specified for the object is used.
out_of_range Indicates when a value exceeds the allowed range of values.

The following program is an illustration of application standard exception class.

PROGRAM 18.13 – The program illustrates the application of overflow_error.

#include <iostream>
#include <stdexcept>
using namespace std;
double Reciprocal(double A) throw (overflow_error)
{ cout<< “Write a small number ”; cin>> A ;
if(A < 0.000001)
throw overflow_error (“Reciprocal is too large.”);
return 1/A ;
}
int main ()
{
try
{
double B =0.0;
cout << “Reciprocal of B = ” << Reciprocal(B) <<endl; }
catch ( exception & x)
{ cout<< “Exception -> ” << x.what() <<endl; }
return 0;
}

The output when the numbered entered is 0.00000001 is given below.

❖ 448 ❖ Programming with C++
Write a small number 0.00000001
Exception -> Reciprocal is too large.
The output, when the number entered is 0.00004, is given below.
Write a small number 0.00004
Reciprocal of B = 25000

The following program illustrates the throwing of invalid_argument of class exception.

PROGRAM 18.14 – Illustrates standard functions of class <exception>.

#include<iostream>
using namespace std;
#include <cmath>
#include <exception>
double Square_root ( int n) throw ( invalid_argument)
{
if ( n < 0)
throw logic_error (“invalid_argument in square root”);
return sqrt (n);}
int main ()
{
int m;
cout<< “Enter an integer number ”;
cin >> m ;
try
{double Square_root (m);
cout<<“The square root of ”<<m<<“ is ”<<Square_root(m)<< endl;}
catch ( exception& exc)
{cout << exc.what()<<endl; }
return 0;
}

The output of first run when m =16 is given below.

Enter an integer number 16
The square root of 16 is 4
The second run with m = –16 gives the following output.
Enter an integer number –16
invalid_argument in square root

18.8 FUNCTION terminate() and set_terminate()

When an exception is thrown and there is no corresponding catch block or is not caught, the
default action is that function terminate() is called which by default calls function abort()
to terminate the program. However, the programmer can have a second option by defining the
 Exception Handling ❖ 449❖
function t e r m i n a t e ( ) and by registering it in the program by the function
set_terminate(). See the following program for illustration.

PROGRAM 18.15 – Illustrates terminate() and set_terminate() functions.

#include<iostream>
#include<stdexcept>
using namespace std;
void Terminate ()
{ cout<< “An exception has occurred.”<<endl;
cout<<“The program is being terminated.”<<endl;
exit (1);}
int main ()
{ set_terminate (Terminate);
try
{
int A;
cout<< “Enter a number less than 60 : “; cin >> A;
if ( A > 60)
throw A;
cout<<“The number entered is “<< A <<endl; }
catch (double)
{ cout<<“Number out of range.”<<endl;}
return 0;
}

The expected output is given below.

Enter a number less than 60 : 70
An exception has occurred.
The program is being terminated.

18.9 FUNCTION unexpected() and set_unexpected()

If a function throws an exception which is not listed in the exception specification, the function
unexpected()is called which by default calls the function terminate(), and which in turn
calls abort() to terminate the program. The function set_unexpected gives a second option to
programmer. Instead of having the default output, the programmer can define his/her own
function unexpected() which will be called when an unexpected exception occurs. The user
defined function unexpected() is registered by making the function an argument of the
function set_unexpected(). One such code is illustrated below.
void Unexpect(){ cout<<“Unexpected called”<<endl;
exit(1);}
❖ 450 ❖ Programming with C++
void F() throw ( ) // the function cannot throw an exception
{
throw 4.0 ;} // but it throws double

int main ()

{set_unexpected(Unexpect);

try
{ F();
}
catch (double)
{ cout<<“double caught”<<endl;}
return 0;
}

18.10 THE auto_ptr CLASS

We have already dealt with dynamic memory allocation through the operator new. The operator
new allocates memory and returns a pointer and delete is used to delete the object and clear the
memory. If the memory allocation is not successful it throws an exception bad_alloc. However,
if the memory has been successfully allocated then it would not throw an exception bad_alloc.
But if an exception happens after the memory has been allocated but before the delete, the
allocated memory would not be recovered. The provision of auto_ptr in C++ tries to solve this
situation. C++ Standard Library contains the class auto_ptr in the header file <memory>. The
class provides the facility to allocate memory to objects through a pointer and when the pointer
goes out of scope or is removed the object that it points to is also removed.

PROGRAM 18.16 – Illustrates auto_ptr for class objects.

# include<iostream>
# include<memory>
using namespace std;
class Rect
{ public:
Rect (){cout<< “Constructor function called.\n”;};
~ Rect(){}
void Func(){ cout <<“It is Rect class function.”<<endl;}
};
int main ()
{
Rect R1;
Rect* ptrR1 = &R1 ;
ptrR1 -> Func();

auto_ptr<Rect> ptr(new Rect), ptr2;

ptr-> Func();
 Exception Handling ❖ 451❖
ptr = ptr2;
ptr2 -> Func();

return 0;
}

The output is given below.

Constructor function called.
It is Rect class function.
Constructor function called.
It is Rect class function.
It is Rect class function.

EXERCISES
1. What is an exception in C++?
2. What are the provisions to deal with exceptions?
3. Explain the try, throw and catch method of dealing with exceptions.
4. What do you understand by exception specification?
5. Give an example of a try block.
6. What should match between a throw and catch so that the exception is handled?
7. What are uses of functions terminate () and set_terminate () functions?
8. When do we use the functions unexpected () and set_unexpected()?
9. Which exception is thrown by operator new?
10. Make a small program to illustrate the working of try, throw and catch mechanism.
11. What is the code for catching all types of exceptions?
12. What types of exceptions the following functions can throw?
(i) void Function ()
(ii) void Function(int) throw( int ,double)
(iii) void Function() throw ()
13. Write a program to illustrate the working of terminate () and set_terminate() functions.
14. Write a program to illustrate the function unexpected() and the function set_unexpectd().
15. Make a program to show that if derived class object is thrown the catch block with base class
type would also work.

Answer:

PROGRAM 18.17 – Illustrates that throw of derived class object can be handled by
a catch block with base class type.

#include<iostream>
using namespace std;
❖ 452 ❖ Programming with C++
class Base // class Base defined
{ };
class Derived : public Base // Derived class defined.
{};
int main ()
{ Base base_obj ; //base_obj is an Object of class Base
Derived D_object; // D_object is object of derived class
int A ;
double B , D ;
char ch = ‘C’;
cout<<“Enter two numbers ”;
cin>> A>>B;
try // try block
{
if (B == 0.0)
throw D_object; // throws an object of Derived class
else
{ D = A/B ;
cout << “D = ” << D <<endl; }} // End of try block
catch (int) // exception handling block for int
{ cout << “The throw \”int \” is caught”<<endl; }
catch(Base) // exception handling block for class Base
{ cout <<“The throw \”Base \” is caught.”<<endl;}
return 0;
}

The expected output is given below.

Enter two numbers 6.0 0.0
The throw “Base” is caught.
The above program illustrates that if the derived class object is thrown in the try block, it can
be handled by the catch block with base class type.

❍❍❍
CHAPTER

19.1 INTRODUCTION

Most of the programs involve some kind of input and output. There are several devices for doing
input to a program, for instance, it may be from keyboard connected to computer, it may from
a file on the hard disc, on CD ROM, floppy disc, from an internet file or from a control device
connected to computer. Similarly the output may be displayed on monitor connected to the
computer, it may be directed to printer, to a file on hard disc/floppy disc/CD ROM or to any
other device connected to computer. Of all these devices keyboard is regarded as standard input
device and monitor is regarded as standard output device. In the previous chapters we have
frequently used cin and cout for performing input and output. These two objects are defined
in the header file <iostream> (input/output stream). The cin is meant to do input from the
standard input device, i.e. keyboard and cout carries out output to standard output device which
is the monitor connected to the computer.
In C++ the input/output ( I/O) operations are carried out by member functions of relevant
template classes. Therefore, the I/O operations are object oriented and are type sensitive. The
I/O of a particular data type is carried out by corresponding member function of the class. If
the compiler cannot find a function for the type of data presented, it will give an error signal.
For fundamental types the overloaded functions for insertion operator (<<) for output, and
extraction operator (>>) for input are already defined in the header files <ostream> and
<istream> respectively and are inherited by the class <iostream> which takes care of both input
and output.
Working with files is an important feature of almost all computer applications. The present
days applications deal with very large numbers of files. For instance, the computers for mobile
service providers may have millions of customer files, the records of calls for more than at least
two months may have to be kept which come to trillions of entries. Similarly the central
computer for a big bank may have to deal with several million account files. The input/output
operations with files are facilitated by header file <fstream> which is derived from <ifstream>,
<ofstream> and <iostream> header files.
The first part of this chapter deals with input/output streams for console operations and the
second part deals with the input/output streams for files.
❖ 454 ❖ Programming with C++

19.2 I/O STREAMS FOR CONSOLE OPERATIONS

In C++ the I/O operations are carried out by streams (Fig.19.1). A stream is a sequence of bytes.
In case of input, the input stream flows from the external device to the main memory of
computer and in case of an output, a stream of bytes flows out of main memory to the output
device. In case of standard input the stream comes from keyboard and for standard output the
stream flows from main memory to the monitor. However, the stream may be redirected from
devices other than keyboard in case of input and can be directed to devices other than monitor
in case of output. The input and output streams are illustrated in Fig.19.1 below.

input stream >> extraction operator

Input
device
Program
-------------

output stream
Output
device

Fig. 19.1: Input/output streams

ios_base

basic_ios basic_streambuf

basic_istream basic_ostream

basic_iostream

Fig. 19.2: Stream classes for input/output with console

Figure19.2 shows the different classes associated with I/O operations and their hierarchy. The
class basic_ios derived from ios_base supports both formatted and unformatted I/O operations.
The class <iostream> that we have used in all the programs is a typedef of basic_iostream for
char. The details of various classes are given below.
ios_base and basic_ios The class ios_base is the base class which has functions common to
both input and output of formatted as well as unformatted data.
The class basic_ios is derived from class ios_base. The classes
basic_istream and basic_ostream are derived from basic_ios. The
typedef of basic_ios, i.e. ios, is used for implementing basic_ios for
using char and wios is used for implementing it for wchar_t.
 Input/Output Streams and Working with Files ❖ 455❖

basic_istream The class basic_istream supports input operations. It is derived from

class basic_ios using virtual inheritance. It is implemented by its
typedef istream for using char. It also defines three functions get(),
getline()and read()besides the other functions. Some of these are
discussed below. It contains the overloaded stream extraction
operator (>>) for C++ fundamental types. The wchart_t is supported
by the typedef wistream.
basic_ostream The class basic_ostream supports all output operations. It is derived
from basic_ios and hence inherits the functions of this class. It
defines the function put() for writing single character and function
write() for output of a number of bytes from memory to output
device. For implementing it for char its typedef ostream is used. For
wchar_t the corresponding typedef is wostream.
basic_iostream The class basic_iostream supports all input/output operations. It is
inherited from basic_istream and basic_ostream through multiple
inheritance. Thus it inherits the functions of basic_ios, basic_istream
and basic_ostream. For its implementation, its typedef iostream is
used for using char. For wchar_t the name is wiostream. The class
iostream is contained in header file <iostream>. The iostream
defines four objects cin, cout , cerr and clog which are discussed
below. For wchar_t the corresponding objects of wiostream are
wcin, wcout, wcerr and wclog.
basic_streambuf The class basic_ios contains the pointer for basic_streambuf . This
class provides buffer interface between data and the memory and
other physical devices. Its typedef streambuf is used for
implementing it for char. It supports low level methods of handling
streams in which negligible formatting is required. The classes
basic_filebuf and basic_strstreambuf are derived from
basic_streambuf . The typedefs for char and wchar_t are streambuf
and wstreambuf respectively.

19.3 PREDEFINED STANDARD I/O STREAMS

The following four stream are defined for standard I/O. The extraction operator (>>) and
insertion operator (<<) are defined for fundamental types. For user defined types the user has to
overload the extraction and insertion operators for a particular class objects.
cin for input from standard input device-keyboard.
cout for output to standard output device-monitor
cerr for standard error output on monitor
clog for buffered error output on monitor.
STANDARD INPUT STREAM
The class istream supports methods for input of formatted as well as unformatted data. Its object
cin along with extraction operator ( >>) is used for the input of fundamental types. For user
❖ 456 ❖ Programming with C++

defined types the extraction operator >> is overloaded for the class of input object. In fact the
operator >> is the right shift operator which is overloaded to do the extraction operation. After
each extraction, the extraction operator sends reference back to cin. This allows cascading
operation and we are able to extract a number of values written one after another and each of
them preceded by extraction operator. For instance, if it is required to extract values of n, m, A
and B we may write the input statement as
cin >> n >> m >> A >> B ;
The reference sent back by extraction operator may also be used in a loop. The following
program illustrates the application of cin.

PROGRAM 19.1 – Illustrates application of cin.

#include<iostream>
using namespace std;
void main()
{ int A ;
double B ;
char ch ;
char Name [10];
cout <<“Enter value of A : ”; cin >>A;
cout <<“Enter the value of B : ”; cin >>B ;
cout <<“Enter the value of ch : ”; cin>>ch ;
cout<<“Enter a short name : ”; cin >> Name ;
cout <<“You have entered the following data.”<<endl;
cout <<“A = ”<<A <<“ , B = ”<<B <<“ , ch = ”<<ch <<“ , Name = ” <<Name <<
endl;
}

The expected output of the program is given below.

Enter value of A : 8
Enter the value of B : 5.5
Enter the value of ch : K
Enter a short name : Madhuri
You have entered the following data.
A = 8 , B = 5.5 , ch = K , Name = Madhuri

STANDARD OUTPUT STREAM

For standard output stream we use the object cout of class ostream along with overloaded
insertion operator <<. The operator << is the left shift binary operator which is overloaded to
do the insertion operation for data of fundamental type. For user defined data the overloading
functions for operator << are defined by the programmer for the class of the objects being dealt.
The return value of insertion operator is the reference to the object of class ostream, which
 Input/Output Streams and Working with Files ❖ 457❖

invoked it i.e., cout. This allows the cascading of stream objects. Thus a number of outputs may
be written in the same output statement as illustrated below.
cout << n << m << A << B;
The following program illustrates the application of cout and insertion operator for different
types of data.

PROGRAM 19.2 – Illustrates output with object cout and insertion operator <<.

#include<iostream>
using namespace std;
void main()
{
int A =4;
double B =2.5;
char ch = ‘T’;
char Name [] = “Mona”;
cout << A<<endl;
cout << B <<endl;
cout << ch<<endl;
cout<< Name <<endl;
cout<<Name<<“ paid rupees “<<B<<“ for “<< A <<“ apples.” <<endl;
}

The expected output is given below. The output is self explanatory.

4
2.5
T
Mona
Mona paid rupees 2.5 for 4 apples.

THE CERR AND CLOG

The cerr and clog are also objects of ostream. The cerr is generally used for handling errors. The
cerr does unbuffered (immediate) output on the standard output device, i.e. monitor. The clog
is also an object of ostream for carrying out output to standard device, however, the output is
buffered. It is also used for error handling.

PROGRAM 19.3 – Illustrates application of cerr and clog objects of class ostream.

#include <iostream>
#include <cmath> //included for mathematical Functions
using namespace std;
int main()
{
❖ 458 ❖ Programming with C++
double x ;
cout<<“Enter a double number: ”; cin>>x;
if ( x<0)
{cout<<“The sqrt argument is negative”<<endl;
cout<< “Enter a positive number.\n”;
exit(0);
}
cout<<sqrt(x)<<endl;

cerr.width(20);
cerr.fill(‘*’);
cerr.precision(7);
cerr<<sqrt(6.0)<<endl;
clog << “The square root of 16 is : ”<<sqrt(16)<<endl;
return 0;
}

In the first trial the number entered is –10. The expected output is given below.
Enter a double number: –10
The sqrt argument is negative
Enter a positive number.

In the second trial the number entered is 12 and the expected output is given below.
Enter a double number: 12
3.4641
*************2.44949
The square root of 16 is : 4

19.4 FUNCTIONS OF <istream> AND <ostream>

Below we discuss the commonly used functions of istream and ostream.

Functions get() and put()
The function get() is a member function of class istream and the function put() is a
member function of class ostream. Both the functions are meant to deal with single character.
Thus get() is used to read single character from an input device and function put() is used
to write a character to output device. The arguments for the two functions are discussed below.
The function get()
The function get() fetches a single character which includes white space characters and eof.
It has to be used with an object of class istream, for example, cin.get(). It may be used with
void argument such as cin.get() or with one argument such as cin.get(char*), with
two arguments or three arguments. All these cases are discussed below.
 Input/Output Streams and Working with Files ❖ 459❖
cin.get();
char ch;
ch = cin.get()
In this case the function get() reads a character and returns the value which is assigned to ch.
Program 19.4 illustrates its application in this form. The function get() with single argument
is illustrated below.
cin.get(char*);
In this case the function assigns the input character to its argument. The code is illustrated
below.
char ch;
cin.get(ch);
Program 19.5 illustrates the application in this form. The function get() with two arguments
is illustrated below.
cin.get(char* buffer, size_t n );
In this form the function get() reads one after another the first n – 1 characters into buffer.
The last character is the NULL( \0) character which marks the end of string and is appended
by the system. The application is illustrated in Program 19.6 below.
cin.get(char* buffer, size_t n, char delim );
In this form of the function, the number of characters to be read into the buffer is limited
either by the number (n – 1) or by the delimiting character delim or EOF whichever occurs first.
On encountering delimiting character the function stops reading. Even delimiting character is
not included. Programs 19.6 and 19.7 illustrate the application of this function.

PROGRAM 19.4 – Illustrates application of cin.get()

#include<iostream>
using namespace std;
int main ()
{
char ch ;
cout <<“Write a character :”;
ch = cin.get();
cout <<“The character ch = ”<< ch << endl;
return 0;
}

The expected output is given below.

Write a character :B
The character ch = B
The following program illustrates the second version cin.get(ch).
❖ 460 ❖ Programming with C++

PROGRAM 19.5 – Illustrates application of function get() in the form cin.get(ch).

#include<iostream>
using namespace std;
int main ()
{
char ch ;
cout <<“Write a character : ”;

cin.get(ch);
cout <<“Character ch = ”<< ch;
cout<<“\n”;
return 0;
}

The expected output is given below.

Write a character : B
Character ch = B
The following program illustrates the function cin.get () with two arguments.
cin.get(char*buffer, size_t n);
In this form the function reads into buffer the characters one by one till the total number
is n – 1 or it is end of file. The nth character is the Null character appended by the system.
PROGRAM 19.6 – Illustrates the function cin.get() with two arguments.

#include <iostream>
using namespace std;
void main()
{
int i =0;
char B[20];
cout<<“Enter a name : ” ;
cin.get (B,6); // reads only 5 -one less than 6
// The last character is ‘\0’appended by system.
cout <<“You have written ; ”<< B <<endl;
}

The expected output is given below. The output shows that 5 characters have been read.
Enter a name : Madhuri
You have written ; Madhu

The following program illustrates the function cin.get() with three arguments, i.e. in the
following form.
 Input/Output Streams and Working with Files ❖ 461❖
cin.get( char* buffer, size_t n, char delim )
Here (n – 1) is the maximum number of characters that may be read. It also includes a
delimiting character, i.e. char delim. If this character is encountered the function stops reading
even though the number (n – 1) is not reached. See Program 19.7 below for illustration.

FUNCTION GETLINE()

The function is used with istream object for reading complete line, for instance, cin.getline().
It can take two or three arguments as illustrated below.
char Buffer [ ]= “Learn C++”;
cin.getline( Buffer, size_t n);
cin.getline( Buffer, size_t n, char delimit);
The application of the function is already demonstrated in Chapter 10 in programs 10.9,
10.10 and 10.11. Also see the following program.

PROGRAM 19.7 – Illustrates the function cin.get() and cin.getline().

#include <iostream>
using namespace std;
void main()
{ char ch = ‘r’; //below ‘r’ is used as a delimiting character.
char C[30] ;
cout << “Enter a name : ”;
cin.getline(C,30);
cout<<“You have entered the name : ”;
cout<<C;
char B[20];
cout<<“\nEnter a name : ” ;
cin.get (B,20, ch ); // reads only up to r
cout <<“You have written ; ”<< B <<endl;
}

The expected output is given below.

Enter a name : Adity Malhotra
You have entered the name : Adity Malhotra
Enter a name : Madhuri
You have written ; Madhu
The output shows that in case of cin. get() reading stopped on encountering the character ‘r’.

FUNCTION IGNORE()

The function is used with input streams. The function is coded as below.
cin.ignore (streamsize n, int delimit)
❖ 462 ❖ Programming with C++

The function reads and discards up to n characters or up to the reading of delimit character
or end of file whichever occurs first. The default value of n is 1. The application of function is
illustrated in Program 10.13 in Chapter 10.

FUNCTION PEEK()

This function is used with input stream. The function takes note of the specified character, i.e.
its argument, but it does not take any action. It is coded as below.
cin.peek(‘D’);
The application is illustrated in Chapter 10 in Program 10.3.

FUNCTION PUTBACK()

It is a member function of istream. It returns the previously read character to input stream. The
code is illustrated below.
char ch = ‘M’;
cin.peek(‘S’);
cin.putback (ch);
According to the above code when the character ‘S’ is noticed it is replaced by M. The
application is illustrated in Programs (10.12 and 10.13) in Chapter 10.

FUNCTION PUT()

The function put is used with an object of class ostream such as cout.put(ch); It puts the
argument ch – a single character to output stream. The following program illustrates its
application.

PROGRAM 19.8 – Illustrates the applications of function put().

#include <iostream>
using namespace std;
int main()
{ cout.put(‘D’).put(‘E’).put(‘L’).put(‘H’).put(‘I’).put(‘\n’);
char* Name = “Calcutta”;
cout<<Name<<endl; // the output is Calcutta
cout.put(*Name); // Output is one character C
cout<<“\n”;
char string [] = “Delhi”;
for ( int i = 0; i<5;i++)
cout.put(string [i]);
cout.put(‘\n’);
cout.put(70)<<endl; // the output is a character F
return 0;
}
 Input/Output Streams and Working with Files ❖ 463❖

The expected output is given below.

DELHI
Calcutta
C
Delhi
F
The function put() writes only single character. In the first output statement the function
is used repeatedly to write DELHI. In the second output statement the object cout along with
insertion operator << are used for writing a string “Calcutta”. When the same is tried by
cout.put() the output is only C of Calcutta. The output Delhi is managed by writing
character by character through a for loop and function cout.put(). The last but one line of
program cout.put(70); results in the output of ‘F’ because according to ASCII the
equivalent value of F is 70 and the function put () writes it as a character.

THE FUNCTION READ() AND WRITE()

The functions read() is a member function of istream and is used with input streams such
as cin.read(). The function write() is a member function of ostream and hence it is used
with output streams such as cout.write(). The function write() does output of a number
of bytes from the character array in the memory without any formatting. Similarly the function
read() does input into the memory some bytes without any formatting. Both the functions
read() and write () take two arguments as illustrated below.
char Name[15];
cin.read(Name ,15);
cout.write (Name, 15)
The following program illustrates the application of the two functions.

PROGRAM 19.9 – Illustrates the function write() and read()

#include <iostream>
using namespace std;

int main()
{
char sentence [] = “Go to school”;

cout.write(sentence , 13).put(‘\n’); // put(‘\n’)is used

// in place of endl
char ch[] = “A”;
cout<<“ch = ”;
cout.write(ch,1)<<endl; // asked to write one byte of ch.

char CH[] = “ABCDEFGHIJC” ;

cout.write(CH, 5)<<endl; // asked to write 5 bytes/characters
❖ 464 ❖ Programming with C++
char Name[9];
cout<< “Type a name ”;
cin.read(Name ,9); // asked to read 9 bytes /char
cout.write( Name,9)<<endl; // asked to write 9 bytes from Name
return 0;
}

The expected output is given below. The output shows that white spaces are also read by function
read().
Go to school
ch = A
ABCDE
Type a name Mona lisa
Mona lisa

19.5 FORMATTED I/O OPERATIONS WITH MANIPULATORS

Formatted I/O operations are needed in several applications such as preparations of lists, tables,
alignment of digits in arithmetic manipulations or for better presentations. In C++ there are a
number of predefined manipulators and class member functions which may be used to control
the format. The user defined functions may also be used for formatting. Thus formatted I/O
may be achieved by following three methods.
(i) By standard manipulators.
(ii) By member functions of class ios.
(iii) By user defined functions.

STANDARD MANIPULATORS
The manipulators are like function-type operators defined in header files<iostream> and
<iomanip>. Some of manipulators are non parameterized while others are parameterised. Table
19.1 given below describes the non-parameterized manipulators such as dec, endl, ends, flush,
hex, oct, and ws. These are defined in header file iostream. The parameterized manipulators given
in Table 19.2 are defined in header file <iomanip>.
Table 19.1 – Non parameterized manipulators

Manipulator Affected stream Description

dec Input/output Decimal conversion base (base 10).
endl Output Insert new line in output and flush an output stream.
ends Output Insert Null character (‘\0’) at end of string.
flush Output Flush an output stream.
hex Input/output Hexadecimal conversion base (base 16).
oct Input/output Octal conversion base (base 8).
ws Input Skip leading white space characters from input stream.
Some of the above manipulators are illustrated in the following program.
 Input/Output Streams and Working with Files ❖ 465❖

PROGRAM 19.10 – Illustrates the application of manipulators hex, oct and dec.

#include <iostream>
using namespace std;
int main ()
{ int n= 183;
cout<<“n in hexadecimal base is = ” <<hex<<n <<endl;
cout<<“n in octal base is = ” <<oct<<n <<endl;
cout<<“n in decimal base is = ” <<dec<<n <<endl;
return 0;
}

The expected output is given below. The number 183 is converted into hexadecimal and octal
numbers.
n in hexadecimal base is = b7
n in octal base is = 267
n in decimal base is = 183

The parameterised manipulators given in Table 19.2 are member functions of header file
<iomanip>.

Table 19.2: Parameterized manipulators, header file <iomanip>

Manipulator Affected stream Description

setbase(int n) I/O It sets conversion base format flag to n.
setfill(int C) output It sets fill character to C .
setiosflags (long F) output Sets (turns on) format flags specified by F.
resetiosflags(long R) output Resets (turns off) format flags specified by R.
setprecision (int P) output Sets precision for floating point numbers to P.
setw(int w) output Sets the field width to w.

The following program illustrates some of the manipulators listed in Table19.2

PROGRAM 19.11 – Illustrates the application of manipulators for formatting.

#include <iostream>
#include <iomanip>
#include <cmath>
using namespace std;
int main()
{ double root;
cout<<setw(6)<<“Number”<<setw(20)<< “root(Scientific)” <<setw(15) <<
“root(normal)\n”<< endl;
❖ 466 ❖ Programming with C++
for ( int i =0; i<6;i++)
{ root = sqrt(i);
cout.setf(ios::left);
cout<<setprecision (8);

cout<<setw(10)<<i<<setw(20)<<setiosflags(ios::scientific) <<root <<

setw(20);
// sets width to 10 for i and 20 for root, sets ios::scientific
cout<<setprecision(4)<<resetiosflags(ios::scientific)<<root<< endl;}
// sets precision to 4, unsets ios::scientific flag
return 0 ;
}

The expected output is given below. The output is self explanatory.

Number root(Scientific) root(normal)

0 0.00000000e+000 0
1 1.00000000e+000 1
2 1.41421356e+000 1.414
3 1.73205081e+000 1.732
4 2.00000000e+000 2
5 2.23606798e+000 2.236

Besides the above mentioned manipulators the class ios has member functions which may
also be used in formatting. Some of these are listed in Table 19.3.
Table 19.3 – ios class functions

Function Description
fill() Sets the character for filling unused field width.
precision() Sets the number of digits for output of floating point numbers.
setf() Sets the flag that controls the output format.
unsetf() Turns off the specified flag.
width() Sets the width of field for output.

A number of the functions listed in Table 19.3 have already been illustrated in Chapter 2
in Programs (2.9 to 2.13).

19.6 FORMATTING BY SETTING FLAGS AND BIT FIELDS

Some formatting operations are accomplished by setting formatting flags through the function
setf(). For example, when the width is set, the number of characters may be less than the set
width. By default the empty spaces will be on the left side. For having the output on left and
empty spaces on right, we may set the formatting flags by the function setf() which has the
following two forms
setf( argument1, argument2);
setf (argument1);
 Input/Output Streams and Working with Files ❖ 467❖

The argument1 is for setting formatting flag and the argument2 is the bit field to which the
formatting flag belongs. In the case of using setf() with only one argument, the argument1 is
used. The different flag setting codes for various formatting settings are listed in Table 19.4. For
un-setting the flag the function unsetf() is used and it takes only one argument. The setting
and un-setting are illustrated below.
The flag setting instructions belong to different bit fields. Therefore, while putting in two
arguments care should be taken that the first argument corresponds to the bit field specified by
second argument. This is illustrated below. For having output in scientific notation one may write
either of the following two codes.
cout.setf(ios::scientific, ios::floatfield);
cout.setf(ios::scientific);
But for un-setting the flag only one argument, i.e. the first one is required. For example the
following code un-sets the above setting.
cout.unsetf(ios::scientific);
Table 19.4 – Arguments of setf(argument1, argument2) and resulting effect.
argument1 argument2 Description
Flags value Related bit-field
ios::dec ios::basefield Conversion to decimal base (10).
ios::hex ios::basefield Conversion to hexadecimal base (16).
ios::oct ios::basefield Conversion to octal base (8).
ios::fixed ios::floatfield Normal float notation.
ios::scientific ios::floatfield Scientific notation.
ios::left ios::adjustfield Output left justified.
ios::right ios::adjustfield Output right justified.
ios::internal ios::adjustfield For internal padding between sign or
base indicator and the number when the
number does not fill full width.
The following program illustrates the application of some of the above flag setting codes.
PROGRAM 19.12 – Illustrates the ios functions fill(), width(), precision(), scientific,
setf ( ios::left) and setf(ios::right), unsetf()
#include <iostream>
using namespace std;
int main()
{
cout.width(20);
cout.fill(‘-’);
cout<<“Good morning!\n”;

/*Good morning! will be written in 20 spaces

the empty spaces will be filled with dash(-) */
❖ 468 ❖ Programming with C++
cout.width(40);
cout.fill(‘*’);
//the output will be written in 40 spaces,
//the empty spaces will be filled with star *.

cout<<“ Welcome to programming in C++!\n”;

int n = 4500600;
cout.width(25);
cout.setf(ios::left); // left justification
cout<<n<<endl;

cout.width(25);
cout.setf(ios::right); // right justification
cout<<n<<endl;
double m = 344.567585435656;
cout. width(5);
cout.precision(10); // precision is set to 10 digits.

cout.setf(ios::scientific); // scientific notation

// This could also be coded as
// cout.setf(ios::scientific, ios::floatfield);
cout<< m<<endl;

cout.unsetf(ios::scientific); // unsetting the scientific

// notation. unsetf()takes only one parameter
cout<<m <<endl;
return 0;
}

The expected output is given below.

------Good morning!
******** Welcome to programming in C++!
4500600******************
******************4500600
3.4456758544e+002
344.5675854

The scientific and normal notation is illustrated by the last two lines of output. The other
function width() sets the field width in which the output should be printed. The default
justification is right. So if nothing is specified about justification, the empty spaces if any would
be on the left side. The justification may be changed to left by the following code so that the
empty spaces are on the right.
cout.setf(ios::left);
 Input/Output Streams and Working with Files ❖ 469❖

FORMATTING FLAGS WITHOUT BIT FIELDS.

The formatting flags for the function setf(), which do not have bit fields are listed in Table 19.5
along with brief descriptions against each.
Table 19.5 – Arguments of setf() which do not have related bit field

Flags Description
ios::showbase Show base indicator on output.
ios::showpoint Show decimal point and trailing zeros.
ios::showpos Show + sign before integers.
ios::skipws Skip white space characters.
ios::unitbuf Use unit buffer size.
ios::uppercase Use uppercase for indicators e and x in scientific and hexadecimal.
The following program illustrates some of the above flag settings.

PROGRAM 19.13 – Illustration of ios::uppercase and ios::showbase

#include<iostream>
using namespace std;
void main()
{ int A = 206;
cout.setf(ios::showbase );

cout.setf(ios::uppercase);

cout<<“Number A is presented in different bases.”<<endl;

cout<< “A in hexadecimal = ”<<hex <<A<<endl;
cout<< “A in octal = ” <<oct<< A<<endl;
cout << “A in decimal = ”<<dec<<A<<endl;
}

The expected output is given below. In the output, the base for hexadecimal and octal is shown
in upper case ( see 0X and 0 in last two lines of the output.)
Number A is presented in different bases.
A in hexadecimal = 0XCE
A in octal = 0316
A in decimal = 206

DISPLAYING TRAILING ZEROS

In normal outputs the trailing zeros are omitted. However, by including the following code the
trailing zeros are also included up to the set precision.
cout.setf(ios::showpoint);
This is illustrated in the following program.
❖ 470 ❖ Programming with C++

PROGRAM 19.14 – Use of ios::showpoint for showing trailing zeros.

#include<iostream>
using namespace std;
void main()
{
double A = 60.35, B= 50.0, C= 36.80, D = 79.450;
cout.setf(ios::showpoint);
cout<< “A = ” <<A<<endl;
cout<< “B = ” <<B<<endl;
cout<< “C = ” <<C<<endl;
cout<< “D = ” <<D<<endl;
}
The output is given below. The default precision is 6 digits, so trailing zeros
up to 6 digits are included in output.
A = 60.3500
B = 50.0000
C = 36.8000
D = 79.4500

WORKING WITH FILES

19.7 INTRODUCTION TO FILES

The present day computers are used to store a very large amount of data. You can imagine the
amount of data stored in railway bookings, in bank transactions of large banks, in computers
managing mobile communications and their billing which may aggregate to trillions of bytes.
All this data is stored in files. In some cases such as in banks or telephone service providers, the
number of files may run into millions. The story does not end with just storage. The data has
to be retrieved also. In case of a travel booking agency, for instance, the customer needs instant
answers to his inquiries. So is the case in banks. In such cases the data should be accessible at
random. Such files are called random access files. Others are sequential files in which data is
stored sequentially. In a sequential file if you wish to find some data you have to start looking
from the beginning till you find your data. In such files also, the programmer can create a
structure of file which helps in fast retrieval of the data.
Files are stored on permanent storage devices such as hard disc, tapes or floppy discs which
are magnetic storage devices and CD ROM and DVD which are optical storage devices. The
floppy disc is relatively small storage device (1.41 Mb). Its main use is for moving data from
one computer to another. Use of floppy discs is gradually decreasing because these are easily
corrupted and there are better options available in the form of portable solid state memory devices
and CD ROM. Some computer companies discourage the provision of floppy disc drive on their
personal computers.
 Input/Output Streams and Working with Files ❖ 471❖

The other memory in a computer is RAM which is a volatile memory. The moment you
switch off the computer all the data on RAM is lost. So it cannot be used for permanent storage,
besides its capacity is also very small. Its main use is to fulfil the memory needs of a running
program.
The above discussion shows that storage and retrieval of data are very important features of
all computer operations. The programs and data are stored in files which are often required to
be created, opened for writing and reading and for updating (reading and writing). C++ has no
restrictions on the form or structure of a file. It may be a free form text file or it may be
formatted. It may contain program and data or only data which may be numeric data or
alphanumeric data. In short a file may be used to store any type of information and there is no
restriction on its form or size if enough memory space is available. It is the creator of file who
decides its contents, form and structure. The computer recognises a storage space in memory
through its name. So a file must also have a distinct name in its directory.
No two files can have the same name in the same directory because it will introduce an
ambiguity. However, you can have extensions to a name to create several related files with same
primary name but with different extensions.

FILE IDENTIFIER OR NAME

A file identifier (name) is a sequence of characters, which may be declared as a sequence of
characters or a string. Besides, the name may consist of a single string of characters or two strings
connected by dot operator. The first string is the primary name and the second string is called
extension. The maximum number of characters in a file name depends on the operating system.
In most of the modern operating systems there is no limit on the numbers of characters in the
name though some compilers take the first 31 characters as significant. However, in MS-DOS
the file name is limited to eight characters and extension is limited to three characters. Below
are some examples of file names.
Student.Grades // primary name Student. Extension - Grades
Employee_data // it is a single name
Employee.salary // primary name Employee, extension salary.

OPERATIONS WITH A FILE

The different file operations are listed below.
(i) Assigning an identifier or name to the file – Every file has to have a distinct name or
identifier in its directory.
(ii) Opening the file – It may involve opening a new file if it does not exist, or opening an
existing file.
(iii) File processing – A file is opened for different purposes i.e. for reading only or for writing
only or for both reading and writing or to append some data or delete some data or
modify some data. The purpose of opening a file is called mode of file or file open
mode.
❖ 472 ❖ Programming with C++

(iv) Detecting errors – Sometimes the intended operation is not carried due to some cause.
The cause must be determined so that it can be rectified. For instance, a new file may
not open due to lack of memory space, etc.
(v) Closing the file – After processing, the file is closed. The closure is carried out by calling
member function close() of the file stream classes which are described below. The
function does not take any argument.

19.8 FILE STREAM CLASSES

In C++ three template stream classes are defined for carrying out various operations with files.
These are
(i) basic_ifstream
(ii) basic_ofstream
(iii) basic_fstream
The typedef for char specializations of these are respectively called ifstream, ofstream and
fstream. These classes are derived from the istream, ostream and iostream as illustrated in Fig.
19.3 below. The files manipulations are done through input stream and output stream. These
streams are governed by the three classes, i.e. ifstream, ofstream and fstream, which are contained
in header file <fstream>. Therefore, any program which deals with files has to include the header
file <fstream>. It is better to include the header <iostream> as well if standard I/O operations
are also desired through the file is contained in <fstream>.
For standard input and output operations we use cin and cout along with extraction and
insertion operators respectively. However, for file input/output operations there are no such
predefined standard streams. The input/output stream are defined by the programmer in the
program itself by creating objects of files <ifstream> for input steam and class <ofstream> for
output stream. The file <fstream> may be used for both input and output streams. The streams
get linked to the particular file if the file name is assigned.

Iostream file
ios
streambuf

istream ostream

iostream

ifstream ofstream
fstream

filebuf
basic-fstream fstream file

Fig. 19.3
 Input/Output Streams and Working with Files ❖ 473❖

ifstream – This class supports input operations from a file to a program. As shown in the
Fig.19.3 the class ifstream is derived by multiple inheritances from class istream and class
basic_fstream which has one constructor with void arguments and another with two arguments
i.e. file_name and open mode. Besides it defines three functions as given below.
(i) bool is_open();
(ii) void open(file_name, mode);
(iii) void close();

The first function is used to test whether the file is open or not. The second function can
open an existing file but does not open a new file by default. The third function is used to close
the file. This function has no arguments. The modes are described in Table 19.6.
ofstream – This class supports output operations from program to file. The class is derived in
multiple inheritance from classes ostream and fstream. In this class also there are two constructors
– one constructor with void arguments and another constructor with two arguments, i.e.
file_name and mode. The default mode is out. The various modes are described in Table 19.6.
The object of this class opens an existing file or opens a new file by default if the file name is
assigned to it. This class also defines three functions as given below.
bool is_open();
void open(file_name, mode);
void close();
fstream – This class supports both input and output operations with files. The class is derived
from class iostream. Any program dealing with I/O with files and standard devices should include
the header files <fstream> and <iostream>. The file <fstream> also defines the three functions
mentioned above in ifstream and ofstream.

19.9 FILE INPUT/OUTPUT STREAMS

As already mentioned above, the file processing is carried out with the help of constructors and
class functions of the three classes ifstream, ofstream and fstream. The class ifstream is used for
input operations, i.e. input from file to program, the class ofstream is used for output operations,
i.e. output from program to file, while fstream is useful for both input operation and output
operation. A file may be opened by following two methods.
(i) Using the constructor function of the concerned class.
(ii) Using the member function open () through an object of the concerned class.
The constructor function constructs the object of the class. The objects of class ifstream and
ofstream may be constructed by the name of the stream. The stream gets linked to the file if
the file name is assigned to it. The following code illustrates the declaration of input stream.

INPUT STREAM DECLARATION

ifstream infile (“Myfile”);

Name Name of input Name of file

of class stream or
class object
❖ 474 ❖ Programming with C++

In the above declaration infile is the name of an object of class ifstream just like cin is
an object of class istream. It is initialized with the file name “Myfile” and this attaches the
stream to the file Myfile. The file name is a string of characters and hence is enclosed in double
quotes. The terms ‘instream’ and ‘outstream’ are with respect to program and not with respect
to the file. Thus here infile is the name of an input stream which extracts data from the file
and feeds it to the program. You can put any other name for this as well. The class ifstream is
used for input files, i.e. for reading the files or for extracting data from files by the program.
Similarly for out stream we create an output stream object of the file ofstream as illustrated
below.

OUTPUT STREAM DECLARATION

An illustration of declaration of output stream is given below.
ofstream outfile (“Myfile”);

Name Name of output Name of file

of class stream
Here outfile is an object of class ofstream and because of initialization with “Myfile”
the output stream gets attached to Myfile. Figure 19.4 illustrates input and output streams for
files.

Input stream for extraction data from file

Program Operating
system

Disc drive

Operating
system
Output stream for inserting data into file

Fig.19.4: Input and output streams for files

If a file does not already exist, the ifstream objects do not open a new file by default with
the given name. However, if a file does not already exist, the ofstream objects open a new file
with the specified name. This is illustrated by the following two programs. In the following
program the ofstream object by name out_to_file is declared and is initialised with the file name
“Myfile”. The stream name out_to_file clearly indicates that data flows from program to the file.
The file does not exist, so it opens a new file with name “Myfile”. However, if the file already
exists this statement will open the file but the existing data in the file will be wiped off. In the
following program we open a file and write two sentences into it.
 Input/Output Streams and Working with Files ❖ 475❖

PROGRAM 19.15 – Illustrates opening a new file for writing in to file.

#include <fstream>
#include <iostream>
using namespace std;
int main ()
{
ofstream out_to_file (“Myfile”); // declaration of output stream

out_to_file <<“Let us learn C++.\n”; // writing into file

out_to_file <<“You need a compiler to practice.\n”;
// writes another sentence in file
out_to_file.close(); // closes the file
return 0;
}

In the following program we open a file for reading its contents. The file was created in
Program 19.15. An input stream by name in_from_file is declared and is initialized by name
Myfile same as used in Program 19.15. The contents are extracted with the help of function
getline(). We are already well acquainted with this function and have used it in standard
input stream as cin.getline(). Here we use it with a user defined input stream as illustrated
below.
in_from_file.getline(str,80);

Name of dot operator Name of Name of Number of

stream function string characters to read

PROGRAM 19.16: Illustrates opening an existing file for reading into program.

#include <fstream>
#include <iostream>
using namespace std;

int main ()
{ char str [80];
ifstream in_from_file (“Myfile”);
in_from_file.getline(str,80);
cout<<str <<endl;
in_from_file.getline(str,80);
cout << str<<endl;
in_from_file.close();
return 0;
}
❖ 476 ❖ Programming with C++

Output consists of following two lines of text which were written into the file in Program
19.15.
Let us learn C++.
You need a compiler to practice.
In the above program we have used the function getline() twice because we have to read
two lines. We could as well use a loop. After reading a line, it is displayed on monitor by the
following output statement.
cout << str<<endl;
Also notice that we also declared the character array str as char str[80]; before using
it. With this declaration the compiler allocates 80 byte of memory for it. The function
getline() reads the line which is stored in this memory and is copied for display by standard
output statement. If a number of lines are to be read it is better to use a while loop because, in
general, we do not know how many lines are there in the file. This is illustrated in Program 19.17.

PROGRAM 19.17 – Illustrates reading a number of lines from a file.

#include <fstream>
#include <iostream>
using namespace std;
int main ()
{char str [80];
ifstream in_from_file (“Myfile”);

while(in_from_file) // while loop

{in_from_file.getline(str,80);
cout<<str <<endl;}
in_from_file.close();
return 0; }

The expected output is given below.

Let us learn C++.
You need a compiler to practice.
In Program 19.15 we observed that if a file does not exist the output stream opens a new
file. But the same thing cannot be done by an input stream. Obviously it cannot read a non-
existing file. See the following program for illustration.

PROGRAM 19.18 – Illustrates that if file does not exist then object of class ifstream does
not open a new file.

#include <fstream>
#include <iostream>
using namespace std;
int main ()
{
ifstream infile (“Myfile”);
 Input/Output Streams and Working with Files ❖ 477❖
if(!infile)
cout<< “The file with name Myfile does not exist.”<<endl;
return 0; }

The expected output is given below.

The file with name Myfile does not exist.

The file input/output may be carried out in the same program. Besides the file may be opened
for reading any number of times. However, for writing you have to be careful because by
opening with default mode will wipe out the existing contents of the file. If the file contents
are to be preserved, then a proper open mode should be specified (see Table 19.6).

PROGRAM 19.19 – Illustrates input from and output to file in the same program.

#include<iostream>
#include<fstream>
using namespace std;
main()
{
char Name[40];
char RegNo[11];
int Age;
ofstream To_file ;
To_file.open(“Student”); // To_file is the name of stream,
// it opens a file with name “Student”
cout<< “Enter name of student: “;
cin.getline(Name , 40);
To_file<<Name<<“\n”; // record Students name in file
cout << “Enter Registration Number of student:”;
cin.getline (RegNo , 10)
To_file<< RegNo<<“\n”; // record the registration number
cout<< “Enter age of student :”;
cin >> Age;
To_file << Age<<“\n”; //record the age of student
To_file.close();
ifstream From_file (“Student”); // input stream from file
From_file >> Name ; // extracting name from file
cout<< “ Output from file is :”<<endl;
cout<<“Name = ”<<Name<<endl;
From_file >>RegNo ; // extracting registration number
❖ 478 ❖ Programming with C++
cout<<“Reg.No = ”<<RegNo<<endl;
cout<<“Age = ”<<Age<<endl; // extracting age of student
From_file.close();
return 0 ;
}

The expected output is given below.

Enter name of student: Madhuri
Enter Registration Number of student:21007
Enter age of student :21
Output from file is :
Name = Madhuri
Reg.No = 21007
Age = 21

19.10 FUNCTIONS is_open(), get() and put() for FILES

In a big program it is sometimes not readily known whether a file is still open or if it was opened
and then closed. Besides the opening and closing of files is quite error prone, so it better to check
if the file is still open or not. The function is_open() is provided in all the three classes ifstream,
ofstream and fstream for this purpose. The return value of function is 1 if the file is open and
0 if it is not open. Its application both for open files and closed files is illustrated in the following
program.
We have already used the function get() with istream object cin. This function reads one
character at a time. The function can also be used with ifstream objects as well. Similarly the
function put () which writes one character at a time can also be used with objects of class
ofstream.
Both the functions, i.e. get() and put(), deal with single character at a time. This characteristic
can be used to count the number of characters read or written or a particular character may be
replaced by another character, etc. Its application is illustrated in the following program.

PROGRAM 19.20 – Illustrates functions get(), put() and is_open().

#include <fstream>
#include <iostream>
using namespace std;
int main ()
{ ofstream outfile(“Myfile”);
outfile<<“I find you have keen interest in C++.\n” ;
outfile <<“You need a compiler to practice.\n”;
outfile.put(‘D’); // write character D
outfile.put(‘E’).put(‘L’).put(‘H’).put(‘I’).put(‘\n’);
outfile.close();
 Input/Output Streams and Working with Files ❖ 479❖
if (outfile.is_open()) // use of is_open
cout<< “The file is open.”<<endl;
else
cout<<“The file is closed.”<<endl;
char kh;
ifstream infile (“Myfile”);
while (infile)
{infile.get(kh); // read the file character by character
cout << kh ;} // display it on monitor
infile.close();

if(!(outfile.is_open()))
cout<< “The file is closed”<<endl;
return 0 ;
}

The expected output is given below. It is self explanatory.

The file is closed.
I find you have keen interest in C++.
You need a compiler to practice.
DELHI
The file is closed

19.11 THE FUNCTION open() AND FILE OPEN MODES

The public member functions open() and close() are provided in each of the three classes
basic_ifstream , basic_ofstream and basic_fstream for opening and closing of files. Both are of
type void. The function close() has no arguments. We have already used this function in the
previous examples.
The function open has two arguments, i.e. the first is the name of file and second is the
mode. For example, if the file is to be opened for reading only the mode is ios::in, for writing
only the mode is ios::out, for appending something at the end of file, the mode is ios::app, etc.
The different modes are listed in Table 19.6 below. The function open() is coded as illustrated
below.
ifstream-object.open ( “Name_of_file”, mode)
Let infile be an object of class ifstream. The function open() may be called in
append mode as below.
infile.open(“Myfile”, ios::app);
❖ 480 ❖ Programming with C++
Table 19.6 – File opening modes

Mode code Description

ios::app Mode for appending at the end of file.
ios::ate Mode for opening at end of file.
ios::binary Mode for opening as binary file.
ios::in Open for reading mode (default for ifstream objects).
ios::nocreate In this mode the open () function fails if file does not exist.
ios::noreplace In this mode the open () function fails if the file exists.
ios::out Mode for opening the file for writing. (default for ofstream objects)
ios::trunc Mode for deleting the contents of the file if it exists.
The following program illustrates the use of ios::app for appending more information at the
end of the file created in Program 19.19.
PROGRAM 19.21 – Illustrates the application of mode ios::app.
#include<iostream>
#include<fstream>
using namespace std;

int main()
{
char Name[40];
char RegNo[11];
int Age;
char Grade [2];

ofstream outfile ;
outfile.open(“Student”, ios::app); // To_file is name of stream
// “Student” is the name of file
cout<< “Enter the grade of student: ”;
cin>> Grade;
outfile << Grade;
outfile.close();
ifstream infile (“Student”); // input stream from file
cout<< “Output from file is :”<<endl;
infile>>Name; // extracting name from file
cout<<“Name = ”<<Name<<endl;
infile >>RegNo ; // extracting registration number
cout<<“Reg.No= ”<<RegNo<<endl;

infile >> Age;

cout<<“Age = ”<<Age<<endl; // extracting age of student
infile>> Grade;
cout << “Grade = ” << Grade<<endl;
infile.close();
return 0;
}
 Input/Output Streams and Working with Files ❖ 481❖

The expected output is given below in which the grade is appended.

Enter the grade of student: A+
Output from file is :
Name = Madhuri
Reg.No= 21007
Age = 21
Grade = A+

19.12 FILE POINTERS

Whenever a modification of file contents is required is, it is necessary to reach the spot where
the modification is desired. A file is sequence of bytes with the first byte as the 0th byte. For a
reaching a particular spot you have to know how many bytes it is away from the beginning of
file or from the end of file in a backward motion. A file pointer points to a position in a file,
which is determined by the offset (number of bytes) from the beginning or from the end. You
would have understood that file pointer is different from the pointers we have used in arrays and
classes in which case the pointer value is the address (the byte number of the start byte) in allocated
memory in RAM. The file pointer is not the address, it is simply an offset in bytes from the
beginning of file or end or from current position. When a file is opened in the read mode the
file pointer is positioned at the beginning of the file. Similarly when file is opened in write mode
the pointer is again at the beginning of file because opening in write mode deletes the existing
contents of the file, so the pointer is positioned at the beginning of file. In case of append mode
(ios::app) the pointer is at the end of file ready to add the additional data. In the append mode
the existing contents remain intact. The Figure 19.5 illustrates the pointer positions for read, write
and append modes.
Open for writing

File start File end

Output pointer

(a) infile.open(“Myfile”, ios::out);

Open for reading

File start File end

Input pointer

(b) infile.open(“Myfile”, ios::in);

Open for appending

File start File end

Output pointer
( c ) infile.open(“Myfile”, ios::app);
Fig.19.5: Position of file pointer when file is opened for writing, reading and append modes
❖ 482 ❖ Programming with C++

Two pointers, i.e. input pointer and output pointer are associated with every file. The input
pointer which is used for reading the file from a particular location is called get pointer. The
output pointer which is used for writing in the file is called put pointer. The four functions listed
in Table 19.7 are used to shift the pointer to a location in the file.
Table 19.7 – Functions for file pointers

Function Header file Description

seekg() <ifstream> It moves the file get pointer to location specified by its arguments
which are discussed below. Used for reading.
seekp() <ofstresm> It moves the file put pointer to specified location by its arguments
which are discussed below. Used for writing.
tellg() <ifstream> It returns the current position of the get pointer.
tellp() <ofstream> It returns the current position of put pointer.

THE FUNCTIONS SEEKG() AND SEEKP()

Both the functions have two arguments. The first is an integer number which specifies the offset
and the second is the seek_dir which specifies the reference position from where the offset
is measured, i.e. from file beginning, from the current position of pointer or from file end. The
codes for the three references are given in Table 19.8 below. The codes for calling the two
functions are illustrated below.
Seekg( int offset, reference_position);
Seekp( int offset, reference_position);

Table 19.8 – Codes for reference positions

Code for reference point Description

ios::beg From the beginning of file.
ios::cur From the current position of file.
ios::end From the end of file.

The Fig.19.6 illustrates of the three references given in Table 19.8.

0 1 2 3 4 5 6

File beginning File end

Offset from beginning = 6 byte Position of file pointer

(a) seekg (6, ios::beg);

File beginning File end

Position of file pointer Offset from end of file (–5)

(b) seekg (–5, ios::end);
 Input/Output Streams and Working with Files ❖ 483❖

File beginning File end

Current position Offset from current position 6 bytes

(c) seekg (6, ios::cur);

Fig. 19.6: Reference points for file pointers

Table 19.9 – Codes for seekp() arguments and the file pointer position

seekp() function option Description of action

infile.seekp(0,ios::beg) Go to the begin of file for writing.
infile.seekp(0,ios::cur) Stay at the current position for writing.
infile.seekp(0,ios::end) Go to the end of file for writing.
infile.seekp(n,ios::beg) Move forward by n bytes from the beginning for writing.
infile.seekp(n,ios::cur) Move forward by n bytes from the current position for writing.
infile.seekp(-n,ios::end) Move back by n bytes from the file end for writing.

The functions tellp() and tellg() return an integer which indicates the current position of the
pointer for writing and reading respectively. The following program illustrates the application
of these functions along with the function seekp(). The comments included at various points
have made the program self explanatory.

PROGRAM 19.22 – Illustrates application of functions seekp(), tellp(), tellg(), ios::in and
ios::app

#include <fstream>
#include <iostream>
using namespace std;
int main ()
{ ofstream outfile(“Mfile”);
outfile<<“Bela! Learn C++ file pointers\n”; // write into file
outfile.close();
char str [80];
ifstream infile (“Mfile”)
infile.getline(str,80); // read the file
cout<<str <<endl<<endl; // display it on monitor
infile.close(); // close the file
infile.open (“Mfile”, ios::in); // open file for reading
int g = infile.tellg(); // tell position of get pointer
cout<<“The pointer is at position g = “<<g<<endl;
// display the pointer position
❖ 484 ❖ Programming with C++
outfile.open(“Mfile”, ios::app); // open for writing at end
outfile<<“which depict position in file”; // add this line
int p = outfile.tellp(); // tell position of put pointer.
cout<<“The pointer is at position p = “<<p<<endl;

infile.getline(str,80); // read first line

cout<<str<<endl;
infile.getline(str,80); // read the next line
cout<<str<<endl;

outfile.seekp(5, ios::beg ); //move 5 bytes from beginning

int n = outfile.tellp(); // tell position of pointer

cout<<“The pointer is at position n = ”<<n<<endl;

outfile.seekp(-3, ios::cur ); // move the back 3 bytes
int m = outfile.tellp(); // now tell the position
cout<<“The pointer is at position m = ”<<m<<endl;
outfile.close();
infile.seekg(–6, ios::end ); // move back 6 from file end
int a = infile.tellg(); // tell pointer position
cout<<“The pointer is at position a = ”<<a<<endl;
outfile.close();
infile.close();
return 0;
}

The expected output is given below.

Bela! Learn C++ file pointers

The pointer is at position g = 0

The pointer is at position p = 60
Bela! Learn C++ file pointers
which depict position in file
The pointer is at position n = 5
The pointer is at position m = 2
The pointer is at position a = 54

19.13 BINARY FILES AND ASCII CHARACTER CODES

We all know that computer stores all types of data in the binary sequences of zeros and ones.
However, the binary forms will differ if we store a digit as a character or store it as its numerical
value that it represents. This is so because value as a character, as per ASCII code, is different from
numerical value that the digit represents. For instance, the number 1242 has the numerical value
twelve hundred forty two. If it is stored as short int it would be allocated two bytes and its binary
representation is 10011011010 which is called binary format stored as illustrated below.
 Input/Output Streams and Working with Files ❖ 485❖

1 0 0 1 1 0 1 1 0 1 0

Fig.19.7 (a): Number 1242 stored in binary

Storing the same number as digit characters takes 4 bytes. One character is allocated 1 byte.
1110 1111 10001 1111

ASCII code ‘1’ = 14, ‘2’ = 15, ‘4’ = 17, ‘2’ = 15

Fig. 19.7 (b): Storing the same number 1242 as digits (characters format)
On a display system (monitor) or a printer it is the characters that are displayed or printed.
So the numerical value stored in the memory as a value of digit is converted into character format
and displayed. Similarly when we enter the numerical value of a variable from keyboard we enter
as a character and the character format of a digit is converted to the numerical value and stored
as value of the variable in the computer memory.

19.14 FUNCTIONS write() AND read( ) FOR FILE OPERATIONS

We have already discussed above the binary format and the character format of representing
the numerical values. The functions write() and read() process the files in the binary format. This
is the format followed for storage in computer memory. The functions get() and put() discussed
earlier in this chapter process the file in character format.
PROGRAM 19.23 – Illustrates functions read() and write() for files operations.

#include <fstream>
#include <iostream>
using namespace std;

int main ()

{ int Bill [5] = { 5, 6, 7, 8, 9};

ofstream outfile (“Myfile”);

for ( int i=0 ; i<5 ; i++)
outfile.write((char*) &Bill, 5);

outfile.close();

ifstream infile (“Myfile”);

infile.read( ( char *) &Bill, 5);

for ( int k =0;k<5 ; k++)

cout << “Bill [“<<k<<“] = ”<< Bill[k]<<“ and size in bytes = ”
<<sizeof(Bill[k])<< endl;
return 0;
}
❖ 486 ❖ Programming with C++

The expected output is as below. The size of elements is 4 bytes, same as for int.
Bill [0] = 5 and size in bytes = 4
Bill [1] = 6 and size in bytes = 4
Bill [2] = 7 and size in bytes = 4
Bill [3] = 8 and size in bytes = 4
Bill [4] = 9 and size in bytes = 4

19.15 FILE OPERATIONS FOR CLASS OBJECTS

The functions write () and read ( ) illustrated above can be used to store object data into file and
read the same from file when required. Since the object data members are stored sequentially,
the object may be treated a single unit occupying a number of bytes equal to its size. The size
of object, which is required in these functions, can be obtained by sizeof( ) function. The input
and output of data is carried out in the form it is stored in the computer memory. However,
the object data consists of the data members only and not the class function members. In the
class program all the objects share the same copy of class function members. So the function
members are not stored with anyone object. Even the type of object is not there when the object
data is stored in a file. This gives rise to a serious problem when objects of several classes are stored
in same file. If the program knows the type of data it is reading from the file it can construct
object from it. Otherwise, the information about object type should be incorporated by
overloading insertion operator. The following program illustrates the write() and read() functions
for writing and reading the objects of single class.

PROGRAM 19.24 – Illustrates file operations with class objects.

#include <fstream>
#include <iostream>
using namespace std;
class Grades
{
private:
char Name [30];
int Reg_No ;
char grade;
public:
void Read()
{cout<<“Enter name of student: ”;
cin>>Name;
cout<< “Enter the registration number: ”;
cin >> Reg_No;
cout<<“Enter the grades : ”;
cin >> grade;}
void Write()
 Input/Output Streams and Working with Files ❖ 487❖
{cout.width(27);
cout.setf(ios::left);
cout<<Name;
cout.width(10);
cout.setf(ios::left);
cout<<Reg_No;
cout.width(2);
cout.setf(ios::left);
cout<<grade<<endl;}
};

int main ()
{
ifstream infile (“STDfile”);
ofstream outfile (“STDfile”);
cout<< “Enter the names, Reg_No and grades \n” ;
Grades Grad [4];
for (int j=0; j<4; j++ )
{Grad[j].Read();
outfile.write( (char*) & Grad[j], sizeof(Grad [j]));
}
outfile.close();
cout<< “\nStudent’s Grade List \n”;
cout << “Name \t\t Reg_No\t Grade”<<endl;
for (int i =0; i<4;i++)
{ infile.read((char*)& Grad[i], sizeof( Grad[i]));
Grad[i].Write();
}
infile.close();
return 0;
}

The program output is given below.

Enter the names, Reg_No and grades
Enter name of student: Sunita
Enter the registration number: 21007
Enter the grades : A
Enter name of student: Madhuri
Enter the registration number: 22007
Enter the grades : B
Enter name of student: Nagacharji
Enter the registration number: 23007
❖ 488 ❖ Programming with C++
Enter the grades : C
Enter name of student: Mamtani
Enter the registration number: 24007
Enter the grades : A

Student’s Grade List

Name Reg_No Grade
Sunita 21007 A
Madhuri 22007 B
Nagacharji 23007 C
Mamtani 24007 A

19.16 RANDOM ACCESS FILES

Often it is required to update the data in a file which may involve insertion or deletion of data
in the middle of file or at ends of a file. A sequential file is a continuous record in which every
byte is occupied. For example, let there be an entry “Mohan has not paid his bills.” If there are
many such entries scattered in the file you have to see the file from the beginning to this particular
entry to get its location. Say the name Mohan is a wrong entry and you want to change the name
to ‘Mohan Katwala. After the correction it will read “Mohan Katwala paid his bill.” The
additional bytes (space) required for the new name have eaten the subject matter and has changed
the entire case. However, the correct correction may be made but in tedious way, i.e. by putting
the contents after the name in a separate file, correcting the name and then appending the
separated part of the file to the name.
A better solution to such problems is to have random access file. In such a file each object
occupies a specified number of bytes. If each object is allocated m bytes then to read nth object
you have to have an offset of (n – 1)*m bytes. You can easily reach the required object. Within
the object space every entry is allotted a specific memory space. Say you may put 50 bytes for
name (limiting the name to 49 characters) etc. Within the allocated space any correction may
be performed without affecting any other entry for that object as well as for the other objects.
Figure 19.8 illustrates a random file.

Object1 Object2 Object3 Objectn

0 m 2m 3m (N–1)m

Fig. 19.8: A random access file

19.17 ERROR HANDLING FUNCTIONS

The stream 1/O oprations like any other computer operation are prone to errors which may be
due to programmer or due to constraint of the hardware such as insufficient memory or due to
user. The errors may be of following types.
 Input/Output Streams and Working with Files ❖ 489❖

(i) Using an invalid or wrong file name.

(ii) The user attempts to read a file which does not exist.
(iii) Attempt to write on a file which is opened for read only.
(iv) An attempt to open a read-only file for writing.
(v) An attempt to open a new file with a name which already exists.
(vi) An attempt to read the file when file pointer is already set at end of file.
(vii) Insufficient disc space.
(viii) Hardware problem.
STREAM STATUS FLAGS
To deal with stream error conditions the class ios_base declares an enumerated type (anum
io_state) with following members. These are explained in Table 19.10.
goodbit = 0x00
badbit = 0x01
eofbit = 0x02
failbit = 0x04

Table 19.10 – Status bit is set according to the condition

Bit Description
goodbit The bit is set when non of other bits are set. It is detected by
cin.good();
badbit It is set when an error results in loss of data. It may be detected
by cin.bad();
eofbit End of the file bit is set for an input stream when end of file is
encountered. It can be detected by cin.eof();
failbit The bit is set when format error occurs in the stream. The data
is not lost in such error. It is detected by cin.fail();

The functions listed in Table 19.11 check the status of the error bits and return values as
dessribed in the table.
Table 19.11 – Error handling functions

Function Action and return value

bad() Returns true if error has occurred, false otherwise.
good() Returns true when no error has occurred, false otherwise.
clear() It clears error states.
eof() Returns true when end of file is detected.
fail() Returns true if an operation, i.e. read or write has failed. The function checks
the bits for the purpose.
rdstate() Returns the stream state data.
❖ 490 ❖ Programming with C++

PROGRAM 19.25 – Illustrates stream status functions.

#include<iostream>
using namespace std;
int main()
{
double D;
cout<< “Initial stream state:”<<endl;
cout<<“cin.bad = ” << cin.bad()<<endl;
cout<<“cin.good = ” << cin.good<<endl;
cout<<“cin.fail= ” << cin.fail()<<endl;
cout<<“cin.eof = ” << cin.eof()<<endl;
cout<< “Enter value of D.”;
cin>> D;
cout<<“User put a string \”Sham\” for D.”
<<“The stream state after input”<<endl;

cout<<“cin.bad = ” << cin.bad()<<endl;

cout<<“cin.good = ” << cin.good<<endl;
cout<<“cin.fail= ” << cin.fail()<<endl;
cout<<“cin.eof = ” << cin.eof()<<endl;
cin. clear();
cout<<“After application of function clear():”<<endl;

cout<<“cin.bad = ” << cin.bad()<<endl;

cout<<“cin.good = ” << cin.good<<endl;
cout<<“cin.fail= ” << cin.fail()<<endl;
cout<<“cin.eof = ” << cin.eof()<<endl;
return 0;
}

The expected output is given below.

Initial stream state:
cin.bad = 0
cin.good = 1
cin.fail= 0
cin.eof = 0
Enter value of D.Sham
User put a string “Sham” for D.The stream state after input
cin.bad = 0
cin.good = 1
cin.fail= 1
cin.eof = 0
After application of function clear():
cin.bad = 0
cin.good = 1
cin.fail= 0
cin.eof = 0
 Input/Output Streams and Working with Files ❖ 491❖

EXERCISES
1. What are the input/output streams?
2. What do you understand by low level I/O and high level I/O ?
3. What are the stream classes for managing I/O with console?
4. To which I/O stream class do the following objects belong?
(i) cout (ii) cin
(iii) cerr (iv) clog
5. What is the difference between the function put() and the function write()?
6. Elaborate the difference between function get() and getline().
7. What is a file pointer? Why can’t we use ordinary pointer in case of files?
8. Which header filed is included for working with parameterized manipulators?
9. For which data-type the function precision () is applicable?
10. What are the left and the right justifications? How are these set?
11. What does function ignore () do?
12. Elaborate the action of the function peek ().
13. What for are the following manipulators used?
(i) oct (ii) hex
(iii) dec
14. How do you declare the file out_stream and file in_stream?
15. Which stream classes are included for carrying out file input and file output?
16. What are the formatting flags? For what actions the following flags are set?
(i) ios::fixed (ii) ios::scientific
(iii) ios::left (iv)) ios::oct
17. What do you understand from the following code lines?
infile.seekp(n,ios::beg)
infile.seekp( n,ios::cur)
infile.seekp(–n,ios::end)

18. What are the file opening modes?

19. What do you understand by following file opening modes?
ios::binary
ios::in
ios::nocreate

20. Show the output for the following statements.

(i) cout.width (20) << “ Ram Chander ”<< endl;
(ii) cout << scientific << 3543.14159 ;
❖ 492 ❖ Programming with C++

(iii) cout << “Good\nMorning”<< “\nKim\n”;

(iv) cout << setw(20) <<setprecision ( 4) << 53426.87566;
(v) cout << hex << 484 << oct<< 400;

21. Show the output for the following statements.

(i) cout.fill(‘*’);
cout.width(10);
cout<< 8546<<endl;
(ii) cout<<“Hello! \t Good morning\n;”
22. What do you understand by following?
(i) failbit
(ii) badbit
(iii) goodbit

23. What do the following function do ?

(i) good() (ii) clear ()
(iii) eof() (iv) fail()

24. What is the structure of a random access file?

25. What is the difference between character format and binary format?

❍❍❍
CHAPTER

20.1 INTRODUCTION TO NAMESPACES

The concept of namespace is an important addition to C++. The need was felt after 1990 when
C++ was gaining popularity and several organizations began offering reusable programs. It is
natural to expect name clash problem if one program is bought from one vendor and the second
from another. Name clash refers to same name being used for different variables. Similar problems
may occur in development of a big program if several programmers are engaged to write code
for different parts of the program. It is quite likely that some names may be common in different
sub-programs though they point to different variables, constants or functions. Naturally when
these sub-programs are put together in a single program, it is not likely that the program will
work or give desired results because of the ambiguities in names. The remedy is (i) either to change
the common names or (ii) to compartmentalize the names under different namespaces. The first
option is a tedious task as well as time consuming. The second option involves putting the names
in different namespaces. The concept of namespace was introduced in C++ in 1995. It is
explained below.
In order to illustrate the concept, let us consider that there are several fruit baskets containing
nearly identical fruits. Say apples are in all of them. If we simply say apples it can refer to apples
contained in any of the baskets. However, if we number the baskets or give them names like
basket1, basket2, basket3, etc. then we can easily identify the apples in basket1 from the apples
in basket3, etc. In C++ code we may refer to apples in different baskets as,
basket1 :: apples ; // this refers to apples in basket1
basket3 :: apples ; //this refers to apples in basket 3
In two or more sub-programs, if apple is a common name of different variables and basket1,
basket2 and baskets3, etc., are the names of namespaces declared in the programs or even in
the different sections of same program, the different sets of names can be easily identified with
the help of scope resolution operator and identifiers for namespaces. Alternatively, if we want to
examine or deal with all the fruits in a basket, we say take basket1, or take basket3, etc. The
equivalent code in C++ would be using namespace NS1 where NS1 is the name of namespace.

20.2 APPLICATION OF NAMESPACES

We have to first declare the namespace at file scope for a program or subprogram. This is similar
❖ 494 ❖ Programming with C++

to giving names to baskets or numbering the baskets as discussed in the above example. The
namespace may be declared as below.
namespace NS1 { Declaration_of_variables ; }
The first word is the keyword namespace, it is followed by the identifier NS1 which is in
turn is followed by braces{} in which the names of a set of variables are declared. For calling the
variable in the program we have to make use of scope resolution operator as illustrated above
in case of apples and baskets. The following program illustrates the concept. We declare some
variables under the namespace S1. Same names are used for different variables under another
namespace S2. The compiler can identify which name belongs to which namespace. Wherever
we use a particular variable we have to first write its namespace identifier followed by scope
resolution operator (::) and then the name of variable. That is also called fully qualified name.

PROGRAM 20.1 – Illustrates application of namespaces.

#include <iostream>
using namespace std;
namespace S1 // name of namespace S1
{int a = 8; //int variable defined under namespace S1.
char ch = ‘X’ ;
double b = 10.5; // variable defined under namespace S1.
char E[] = “John!”; }
namespace S2
{double a = 5.4; // ‘a’ is now double in namespace S2
char ch = ‘Y’; //ch is now ‘Y’ was ‘X’ under S1
int E = 5; // E is now int was string in S1
char b[] = “Go to school”;} // b is now string was double in S1
void main ()
{double A = S1::a * S2::a; //the scope resolution identifies
// whether it belongs to S1 or S2
double B = S1:: b + S2::E ;
cout << “A = ” << A <<“ , ” << “B = “<< B << endl;
cout << S1::ch <<“ , ” <<S2 :: ch<<endl;
cout << S1::E <<“ ”<< S2::b<< endl;
}

The expected output is as below.

A = 43.2, B = 15.5
X,Y
John! Go to school
We have already been using the declaration using namespace std; for simplifying the
writing of std::cout to simply cout, etc. The C++ header files and functions are written under
the namespace std.
 Namespaces and Preprocessor Directives ❖ 495❖

20.3 DIRECTIVES using AND using namespace

In the above program we have used scope resolution operator (::) preceded by the identifier for
namespace to access the desired variable. This may also be done by using the keyword using
followed by namespace identifier and scope resolution operator to reach a single name whether
it is for variable or function. If several names declared under one namespace are to be used in
the section of a program then we may use the following directive.
using namespace identifier_of_namespace;
The following program illustrates the above option.

PROGRAM 20.2 – Illustrates application of namespaces with using namespace directive.

#include <iostream>
using namespace std;
namespace doll // Here doll is identifier for namespace
{ int a = 8;
char E[] = “John!”; }
namespace ball // Here ball is identifier for namespace
{double b = 4.5;
char E[] = “Go to school”;}
void main ()
{using namespace doll; // application of using directive
cout << a*a<< endl; // now scope resolution not needed
//for variable of namespace doll but
//it is needed for variables of namespace ball
cout << E <<“ ”<< ball:: E<<endl;
// use of scope resolution operator for E
}

The expected output is given below.

64
John! Go to school
The following program illustrates application of keyword using for name of one variable.

PROGRAM 20.3 – Illustrates the use of keyword using.

#include <iostream>
using namespace std;

namespace NS1
{ int n = 3;
float m = 2.5;}
namespace NS2
❖ 496 ❖ Programming with C++
{float n = 4.0;
int m = 2 ;}
namespace NS1
{
int k = 2;
double R = n*m*k;}
namespace NS2
{double k = 3.0 ;
double R = n*m*k;
}
int main()
{
using NS1::R; // Application of using
cout<< “R of NS1 = ”<< R <<endl;
using NS1::k; // Application of using
cout << “k of NS1 = ”<< k << endl;
cout<< “m of NS2 = ” << NS2::m<< endl;
using NS1::m; // Application of using
cout << “m of NS1 = ”<< m <<endl;
cout << “R of NS2 = ” << NS2::R<< endl;
return 0 ;
}

The expected output is given below.

R of NS1 = 15
k of NS1 = 2
m of NS2 = 2
m of NS1 = 2.5
R of NS2 = 24

20.4 NAMESPACE ALIASES

Aliases for namespace identifiers may be used wherever it is convenient. With short names there
is greater likelihood of name clash. So it is better to have a big name to start with and then
declare a short name as alias for frequent use. This is illustrated in the following program.

PROGRAM 20.4 – Illustrates aliases for namespaces.

#include <iostream>
using namespace std;
namespace myspace_program
{ int n = 3;
 Namespaces and Preprocessor Directives ❖ 497❖
float m = 2.5;
int k = 2;

double R = n*m*k;
}
namespace Myspace
{float n = 4.0;
int m = 2 ;
double k = 3.0 ;
double R = n*m*k;
}
int main()
{
namespace NS1= myspace_program ;
//NS1 is alias for myspace_program
namespace NS2 = Myspace ; // NS2 is alias for Myspace

cout<< “R in myspace_program= ”<< NS1:: R << “ , \tn in NS1 = ” <<NS1::n<<endl

cout <<“R in Myspace = ” <<NS2::R<< “ , \tn in NS2 = ” << NS2::n <<endl;
cout<<“m in myspace_program = ” <<NS1::m<< “ , \tm in NS2 = ” <<NS2::m <<
endl;
return 0 ;
}

The expected output is given below.

R in myspace_program = 15 , n in NS1 = 3
R in Myspace = 24 , n in NS2 = 4
m in myspace_program = 2.5 , m in NS2 = 2
The following program is yet another instance of aliases for namespaces.

PROGRAM 20.5 – Another illustration on aliases to namespaces.

#include <iostream>
using namespace std;

namespace NS1
{ int n = 3;
float m = 2.5;}
namespace NS2
{float n = 4.0;
int m = 2 ;}

namespace NS1
{

int k = 2;
double R = n*m*k;}
namespace NS2
{ double k = 3.0 ;
❖ 498 ❖ Programming with C++
double R = n*m*k;
}
namespace Myspace = NS1;
namespace Space = NS2;
int main()
{
cout<< “R of Myspace = ”<< Myspace:: R <<endl;
cout << “k of Myspace = ”<< Myspace::k << endl;
cout << “k of namespace NS2 = ” << NS2::k << endl;
cout<< “m of Space = ” << Space::m<< endl;
cout << “m of Myspace = ”<< Myspace::m <<endl;
cout << “R of Space = ” <<Space::R<< endl;
return 0 ;
}

The output of the program is given below.

R of Myspace = 15
k of Myspace = 2
k of namespace NS2 = 3
m of Space = 2
m of Myspace = 2.5
R of Space = 24

20.5 EXTENSION OF NAMESPACES

The namespaces may be defined in parts in the same program or may span over several files. The
following program is an illustration of extending namespaces in same program.

PROGRAM 20.6 – Illustrates extension of namespaces.

#include <iostream>
using namespace std;
namespace NS1
{ int n = 3;
float m = 2.5;}
namespace NS2
{ float n = 4.0;
int m = 2 ;}
namespace NS1 // using NS1 again
{ int k = 2;
double R = n*m*k;}
 Namespaces and Preprocessor Directives ❖ 499❖
namespace NS2 // Using NS2 again
{double k = 3.0 ;
double R = n*m*k; }

int main()
{ cout<< “R of NS1 = ”<< NS1:: R <<endl;
cout << “k of NS1 = ”<< NS1::k << endl;
cout<< “m of NS2 = ” << NS2::m<< endl;
cout << “m of NS1 = ”<< NS1 ::m <<endl;
cout << “R of NS2 = ” << NS2::R<< endl;
return 0 ;
}

The expected output is given below.

R of NS1 = 15
k of NS1 =2
m of NS2 =2
m of NS1 = 2.5
R of NS2 = 24

20.6 NESTING OF NAMESPACES

The namespaces may be nested, i.e. a namespace is declared inside another namespace. For instance,
the following is an illustration of nested namespaces.
namespace NS1
{ namespace NS2 //namespace NS2 is declared inside NS1
{int x = 2;}}
With above declaration the variable x may be accessed given below.
NS1::NS2::x;
The following program illustrates the nested namespaces.

PROGRAM 20.7 – Illustrates nested namespaces.

#include <iostream>
using namespace std;
namespace NS1
{ int n = 3;
float m = 2.5;
namespace NS2 // NS2 declared inside NS1
{ float n = 4.0;
int m = 2 ;}}
namespace NS3
{ int k = 2;
double R = NS1::n*NS2::m*NS3::k; // using variables of
// different namespaces
❖ 500 ❖ Programming with C++
namespace NS4 // NS4 declared inside NS3
{double k = 3.0 ; }}

int main()
{ cout<< “R of NS1 = ”<< NS3:: R <<endl;
cout << “k of NS4 = ”<< NS3::NS4 ::k << endl;
cout<< “m of NS2 = ” << NS1::NS2::m<< endl;
cout << “m of NS1 = ”<< NS1 ::m <<endl;
return 0 ;
}

The expected output is given below.

R of NS1 = 12
k of NS4 =3
m of NS2 =2
m of NS1 = 2.5

20.7 THE namespace std

The various header files and names in C++ are now declared under the namespace std. As
discussed above, the namespaces, in general, can be modified by additional declarations and
definitions, however, this does not to apply to the namespace std. The namespace std cannot
be modified. According to earlier convention the header files are written with the extension (.h)
such as <iostream.h>. Now, under namespace std, the same is written as <iostream>, however,
you have to include the statement using namespace std; in the program in global scope, i.e. above
main().
The standard C header files which were earlier written with extension such as (<xxxx.h> are
now written as <cxxxx> under the namespace std. The extension h is omitted and c is added
before the name. For example, the name <assert.h> is now written as <cassert>. The compilers
at present support both the conventions. More examples are given below.
Table 20.1
Earlier name Name under namespace std
Examples of some C++ header files
<algorithm.h> <algorithm>
<bitset.h> <bitset>
<iomanip> <iomanip>
<set.h> <set>
Examples of some ANSI C header files
<assert.h> <cassert>
<float.h> <cfloat>
<limits.h> <climit>
<math.h> <cmath>
<stdio.h> <cstdio>
<stdlib.h> <cstdlib>
<time.h> <ctime>
 Namespaces and Preprocessor Directives ❖ 501❖

20.8 PREPROCESSOR DIRECTIVES

There are a number of preprocessor directives which C++ has inherited from C. These may be
used for defining symbolic constants, macros and conditional directives. We are already well
acquainted with #include and #define. A list of preprocessor directives is given in
Table 20.2. All the preprocessor directives are preceded by the symbol #. For many of these, there
are equivalent replacements in C++. For instance, an inline function does the same as a macro,
i.e. it substitutes the code wherever the name of function appears. In fact, inline function is a
better option than a macro because there is static type checking while a macro bypasses the type
check. Similarly const in C++ does the same as a constant defined by directive #define.
Table 20.2 – Preprocessor directives and operators
Preprocessor directive Description
# # ignored if single on a line. It is Null directive.
# define Used for defining a symbolic constant or a macro.
## Used with # define, the operator concatenates two items.
#error Stops compilation and displays error message.
# include Includes the files names in <> brackets or in double quotes “ ” in
source code. We have used this directive in all the programs.
#line Allows to specify the line number in a file. It is coded as #line
Number_line<“Name_of_file”>.
#pragma Directive specific to implementation.
#undef Undefine the previously defined.

# DEFINE

The preprocessor directive #define is used to create macros and symbolic constants. The directive
#define may be used as illustrated below.
#define Pi 3.14159
Note that there no semicolon at the end of line and there is a space between define and Pi
and between Pi and the value 3.14159. In the program using the above code, wherever Pi occurs,
the value 3.14159 would be substituted. In fact, the above declaration is equivalent to the
following statement.
const double Pi = 3.14159 ;
Macros may also be created by #define. For instance, see the following code.
#define max(x,y) (x >y ? x : y)

Pre-processor Space Function name Space Definition

and arguments
In the above declaration the directive #define creates a macro — a function to obtain the
greater of the given two values. The values may be integers or floating point numbers, or
characters. There is no type checking. Note that in the definistion no space is allowed between
max and the left bracket ‘(‘. Its application is illustrated in the following program.
❖ 502 ❖ Programming with C++

PROGRAM 20.8 – Illustrates preprocessor directive #define.

#include<iostream>
using namespace std;

# define max(x,y) (x >y ? x : y) // definition of macro

int main()
{
float x,y;
char ch, kh;
int A, B;

cout<<“Write two integers ” ; cin >> A >> B;

cout << “Enter two floating point numbers ”;cin >>x>>y;
cout <<“Enter Two characters ”; cin >> ch>>kh;

cout << “Greater of the two int = ” <<max (A ,B ) << endl;

cout << “Greater of the two floats = ” << max(x,y)<<endl;
cout<< “Greater of two characters is = ” << max (ch,kh)<<endl;

return 0;
}

The expected output of the program is given below.

Write two integers 4 8
Enter two floating point numbers 7.7 9.6
Enter Two characters S H
Greater of the two int = 8
Greater of the two floats = 9.6
Greater of two characters is = S

The following program is another illustration of defining a macro by preprocessor directive #

define. None that both length and width are enclosed in brackets. This is to eliminate error that
may occure if, for instance, length = 2+3 and width = 2+2 in which case it would compute as
2+3* 2+2 = 10
instead of 20. This is be cause (*) has higher precedence than +.
PROGRAM 20.9 – Illustrates another function definition with #define.
#include<iostream>
using namespace std;

# define area ((length)*(width))

// the macro defines area of a rectangle.
int main()
{
int length, width;
cout<<“Write the length & width of a rectangle: ” ;
cin >> width >> length;
cout <<“Area = ”<< area << endl;
 Namespaces and Preprocessor Directives ❖ 503❖
return 0;
}

The expected output is as below.

Write the length & width of a rectangle: 60 20
Area = 1200

OPERATORS # AND ##
The single # in its own line is neglected by compiler. Thus,
#
is of no consequence. When used with a variable such as #x in a #define statement, the #
operator is replacement token, i.e. it replaces the value wherever x appears. The replacement is
converted into a string with double quotes. For example, see the following code.
#define F(x) cout << “Good Morning” #x
When function F(x) is called as F(Babu) the above code is equivalent to the following
statement.
cout << “Good Morning Babu”;
The preprocessor operator ## is used along with #define, it puts together (concatenates)
two tokens. For example the following code.
#define concat(m , n) m ## n
would give output as mn. See the following program for illustration.
PROGRAM 20.10 – Illustrates preprocessor operator ##.
#include<iostream>
using namespace std;

#define Funct(y) cout << “Good Morning ” #y<<endl;

# define CONCAT(a, b) a ## b

void main()
{
Funct (Aparna!);
cout << CONCAT (5, 7 )<< endl;
cout << CONCAT (“O”,”K”)<< endl;
cout << CONCAT ( “Good”, “ Morning John!” )<< endl;
}

The expected output is given below.

Good Morning Aparna!
57
OK
Good Morning John!
❖ 504 ❖ Programming with C++

#UNDEF
The preprocessor #undef is used to discard or undefine the symbolic constants and macros
created by #define. Thus the scope of macros and symbolic constants starts from the definition
to the point where it is undefined by #undef or if not undefined to the end of file. There is no
restriction on redefining the same macro or symbolic constant by #define. It is illustrated in
the following program.

PROGRAM 20.11 – Illustrates #define and #undef

#include<iostream>
#define Length 40 // First define should be before main()
using namespace std;
int main()
{ cout << “Initial length = ” << Length << endl;
#undef Length
#define Length 60
cout << “New length = ”<< Length << endl;
return 0; }

The expected output is given below

Initial length = 40
New length = 60

20.9 CONDITIONAL PREPROCESSOR DIRECTIVES

The conditional preprocessor directives are provided to control the action of preprocessors and
to control the compilation of the program. These are listed in Table 20.3 with the description
in each case. Note that if an #if is used it ends with #endif. Program 20.11 illustrates the
application of some of these.
Table 20.3 – Conditional preprocessor directives

Preprocessor directive Description

#if Conditional if directive. The #if is followed by # endif
#elif else if conditional directive
#endif end of if
#else else
#ifdef if defined
#ifndef if not defined

Some of the above directives are illustrated below.

 Namespaces and Preprocessor Directives ❖ 505❖
PROGRAM 20.12 – Illustrates application of #if, #elif, #define and #undef.

#include<iostream>
using namespace std;
#define width 40
#define area (length*width)
#if 0
#if length > 2*width
#undef width // discard the previous definition of width
#define width length/2
#elif length < width // Similar to else if
#endif
#undef width
#define width length/4 //redefinition. No semicolon at end
#endif
int main()
{ int length ;

cout<<“Write the length of a rectangle: ” ;

cin >> length;

cout << “Length = ” <<length<< “\t Width = ” << width<<endl;
cout << “area = ” << area << endl;

return 0; }

The expected output is given below. The code between #if 0 and #endif is not compliled.
This technique is used for debugging programs.
Write the length of a rectangle: 100
Length = 100 Width = 40
area = 4000

PROGRAM 20.13 – Illustrates application of #error, #if, #endif, #else

#include<iostream>
using namespace std;

#define area (length*width)

#define length 100

#if (length>50)
#error 1- Not according to plan
#else

#endif

int main()
{
int width;
❖ 506 ❖ Programming with C++
cout<<“Write the width of a rectangle: ” ;
cin >> width;

cout << “Area= ” <<width*length<<endl;

return 0;
}

First run with length = 10

Write the width of a rectangle: 5
Area= 50
Second run when length =100.
#error : 1- Not according to plan

ASSERT ()
The function assert() is defined in header file <cassert>. It is used to test a value. For
instance one may use it as below.
assert (k>0);
With such a statement if k happens to be 0 or less than 0 the program will be aborted and
the error message will identify the line in which the error occurred. It is illustrated in the
following program.

PROGRAM 20.14 – Illustrates application of assert() function.

#include<iostream>
#include <cassert>

using namespace std;

void main()
{
int A, B;
cout<< “Enter two integers ”; cin >> A>>B;
assert(B > 0);

double C = A/B;
cout << “C = ” << C <<endl;
}

The output, when program was tested on Visual C++6 is given below.
Enter two integers 5 0
Assertion failed: B>0, file C:\Documents and
Settings\Administrator\Desktop\Prep
ro.Assert.cpp, line 7

The following program illustrates other preprocessor functions.

 Namespaces and Preprocessor Directives ❖ 507❖

20.10 PREDEFINED MACROS

A list of six predefined macros is given in Table 20.4. The table also explains the expected output
of these macros. Program 20.15 illustrates the application of some of them.
Table 20.4 – Predefined macros

Name of macro Description

_cplusplus C++ program.
_DATE_ Date month day year (mmm dd yyyy)
mmm = three character for name, i.e. Oct.
_FILE_ Name of source code file.
_LINE_ The line number.
_STDC_ ANSI compliant.
_TIME_ Compilation time ( hh: mm : ss).

The following program illustrates some of the predefined macros.

PROGRAM 20.15 –Illustrates pre-defined macros _LINE_, _TIME_, _DATE_ ,
_FILE_ and _cplusplus
#include <iostream>
using namespace std;
int main()
{
int sum = 0;
for ( int i = 0 ; i<6; i++)
sum +=i;
cout <<“sum = ” << sum<<endl;
cout << “Line No = ” << __LINE__ <<endl;
// double underscore on each side
cout << “Date of writing the program ”<< __DATE__<<endl;
cout << “The time of compilation is ”<<__TIME__<< endl;
cout <<“File No = ”<< __FILE__ << endl;
cout<< “Value of cplusplus ” << __cplusplus <<endl;
return 0;
}

The expected output obtained on Microsoft Visual C++ 6 is as under.

sum = 15
Line No = 9
Date of writing the program Oct 1 2006
The time of compilation is 19:27:31
File No = F:\C++ Prog\c++ All Programmes\Predefined.cpp
Value of cplusplus 1
❖ 508 ❖ Programming with C++

EXERCISES
1. What is a macro?
2. How do you define a macro and undefined a macro?
3. What is a symbolic constant?
4. What does the macro assert do? Make a small program to illustrate its application.
5. Define a macro to find the greater of the two numbers.
6. Define a macro to find greatest of three numbers entered by a user.
7. Define a macro to determine area of a triangle when three sides are specified.
8. Write the expected output from the following program.

#include <iostream>
using namespace std;
void main()
{
cout<< _TIME_<<endl;
cout << _DATE_<< endl;
cout<< _LINE_ <<endl;
return 0;
}

❍❍❍
CHAPTER

21.1 INTRODUCTION

Standard Template Library (STL) is an excellent collection of templatised reuseable software

which deals with data structures and operations most commonly used by programmers in many
diverse applications. The concept was developed by Alxender Stepanov and Meng Lee of
Hewlett-Packard. The collection has contributions by many other researchers and experts in C++
programming. It is now a part of C++ Standard Library.
Use of STL components can save considerable time and effort of programmers, besides
assuring a degree of efficiency and accuracy to the application programs because of their being
based on well tested pieces of software written by experts. STL is a big collection of programs,
however, the following three are its main components.
(i) Containers
(ii) Iterators
(iii) Algorithms
Containers are objects which store other objects which may be data of fundamental type or
user defined type. The storage, retrieval and various operational functions to manipulate data are
based on template classes, so containers allow storage and manipulation of any type of data.
Iterators are like pointers, however, iterators being based on template classes allow their use
for diverse data types. The main function of iterators is to provide access to elements of containers
and thus can be used to traverse the containers as well as for operation on any one element of a
container. There are different types of iterators depending on their capability and hence their use
is accordingly limited. All the containers do not support all types of iterators. The details are
discussed below.
Algorithms are pieces of software which are used to manipulate the data in containers with
help of iterators. The algorithms are not only for STL containers and hence can be used for
operations on other types of containers like arrays, strings or variables in general. In this chapter
all the topics mentioned above are introduced, while details on sequence containers, associative
containers and algorithms are dealt separately in Chapters 22, 23 and 25 respectively.
❖ 510 ❖ Programming with C++

21.2 THE CONTAINERS

The containers are categorised into following three main groups, i.e. (i) sequence containers, (ii)
associative containers and (iii) container adapters as illustrated in the following diagram. These
are further specialised into containers with special functions, so a programmer may choose the
one best suited to the application on hand. Besides the above mentioned three containers there
are three near containers. These are C++ strings, bit sets and valarray. The near containers have
some functions similar to other containers. Figure (21.1) shows the relationships of various
container classes.
Containers

Sequence containers Associative containers Container adapters

vector list deque set multiset map multimap stack queue priority-
queue
Fig. 21.1: Sequence containers, associative containers and adapters
Sequence containers and associative containers are collectively called first class containers
because they store as well as have the capability to manipulate data with the help of iterators and
have sets of predefined functions. The container adapters do not support iterators and hence
cannot manipulate data. Adapter classes support only few functions which are mainly for pushing
in or for removing data from front or back-end of the sequence.
Sequence containers are in fact linear sequences of data. The individual data element is linked
to other elements by way of its position in the sequence. They allow insertion and deletion of
data in the sequence as well as can allocate to themselves additional memory if the number of
elements increases beyond the capacity (memory) initially allocated. In this respect they are
dynamic, unlike C- arrays which are of fixed length right from the point they are declared. The
sequence containers are of three types, i.e. (i) vector, (ii) deque (also called double ended queue
and pronounced as ‘deck’) and (iii) list. Each has speciality of being efficient in some aspects over
the other two. For instance vector allows fast random access to an element as well as addition
and deletion of elements at the back-end only, but is slow in insertion or deletion in the middle.
The list is fast in insertion and deletion in the middle and at front-end but does not support
random access. The list supports only bidirectional iterators, therefore, are slow to access an
element at random. Similarly deque is fast at random access as well as in insertion or deletion
at both ends, but slow in insertion or deletion in the middle. The deque also supports random
iterators. The individual data element may be accessed by iterators and can be manipulated. The
salient features of each of these are given in Table 20.1 below.
 Standard Template Library ❖ 511❖
Table 21.1 – Header files and brief description of containers
For details on sequence containers and associative containers see Chapters 22 &23.

Container Header File Brief Description

vector <vector> The container provides direct access to any element.Rapid addition
and deletion at back. For the functions and operators supported by
vector class see Chapter 22.
list <list> Rapid insertion and deletion at any location in the list. For the list of
functions and operators supported by list class see Chapter 22.
deque <deque> Allows direct access to any element of deque container.Rapid addition
and deletions at both ends of the queue.For list of functions and
operators supported by deque see Chapter 22 .
set <set> Only unique values are allowed, duplicate copies ignored if inserted.
Values are sorted in specified order such as increasing order or
decreasing order, etc.(See Chapter 23)
multiset <set> Multiple copies of a value allowed.The elements are sorted in order
as specified like in sets.
map <map> The container stores pair of values for each element, the first value is
the key and the second is the associated value. The elements are
sorted according to key and only unique keys are allowed. For details
see Chapter 23.
multimap <map> The container stores pair of values for each element, the first value is
the key and second is the associated value. The elements are sorted
according to key. Multiple keys are allowed. (See Chapter 23 for
details).
stack <stack> Stack of objects has input/output order as first in last out order (FILO)
or in other words last in first out. For more details see Sect.21.6 of
this chapter.
queue <queue> A queue of objects. The input/output order is first in first out (FIFO).
For more details see Section 21.6.2 below.
priority queue <queue> A queue of objects. The element of highest priority is at top of queue
and is first out. The details are given in this chapter in Section 21.6
below.

21.3 ITERATORS

Iterators are like pointers, they are used to access the elements of container classes. They are also
used to traverse from one element to another of the container by increment and decrement
operators. The iterators are of following types. However, all of them are not supported by all
the container classes.
❖ 512 ❖ Programming with C++
Iterator Description
input_iterator These can be used to read values. These can be incremented, i.e.
can move only forward. Other operators supported by
input_iterator are given in Table 21.2.
output_iterator They can be used to write and can be incremented and
dereferenced.
forward_iterator They can do the functions of input as well as output iterators.
However, these move in forward direction only. These support
dereferencing and assignment operator.
bidirectional_iterator These can be used to read or write. These can have forward or
backward motion (can be incremented as well as decremented) and
can be dereferenced.
random_iterator These may be used for random access to an element, can be used
for reading as well as writing. Random iterators encompass all the
characteristics of all other iterators.

DECLARATION OF AN ITERATOR
The declaration of iterators for a vector <int> is illustrated below.
Scope resolution

vector < int > :: iterator Iter ;

class template type of Keyword Name of iterator

name vector elements
Iter = V1.begin(); // assigning a value to iterator Iter
// V1 is name of vector.
Fig. 21.2: Declaration and assignment of an iterator for vector class.

The declaration and assignment may be combined as given below.

vector <int> :: iterator Iter = V1.begin();
The value of the element to which the iterator is pointing may be obtained by dereference
operator (*). For instance, the value of first element of the vector V1 from the above declaration
is given by *Iter and value of second element is *(++Iter).

OPERATORS SUPPORTED BY DIFFERENT ITERATORS

There are five types of iterators as already discussed above. Each type of iterator supports a set
of operators as listed in Table 21.2.
 Standard Template Library ❖ 513❖
Table 21.2 – Operators supported by different iterators
Input Output Forward Bidirectional Random Description of
interator interator interator interator interator operator
++itr ++itr ++itr ++itr ++itr pre-increment
itr++ itr++ itr++ itr++ itr++ post-increment
*itr *itr *itr *itr Value = *itr
read only See Program 21.1
*itr *itr *itr *itr *itr = Value
write only See Program 21.2
itr1 == itr2 itr1 == itr2 itr1 == itr2 itr1 == itr2 comparison for equality
itr1 != itr2 itr1 != itr2 itr1 != itr2 itr1 != itr2 comparison for inequality
itr1 = itr2 assignment
––itr ––itr pre-decrement
itr–– itr–– post-decrement
itr + k increase by int k
itr – k decrease by int k
itr += k increase by int k
and assign
itr –= k decrease by int k
and assign
[k] subscript operator
itr1 < itr2 comparison less than
itr1>itr2 greater than
itr1<=itr2 less than or equal to
itr1>=itr2 greater than or equal to

21.4 SEQUENCE CONTAINERS

Some of the features of containers have already been discussed in the Table 21.1. The sequence
containers comprise three containers which are vector, list and deque (double ended queue). Here
we take illustrative examples of each of the three, for more details see Chapter 22.

THE VECTOR
The vectors are dynamic arrays. The number of elements may be changed during the program
execution. Here a few features are illustrated. The following program gives an illustration of
vector, iterators and functions like push_back(), begin(), end(), etc. For other functions
and operators supported by vector class see Chapter 22.
❖ 514 ❖ Programming with C++
PROGRAM 21.1 – Illustrates application of vectors and input iterator of istream.

#include<iostream>
#include<vector>
# include <iterator>
using namespace std;
vector<int> V; // Declaration of a vector with int elements
int main()
{
cout<<“Enter 5 integers one after another.”<<endl;
istream_iterator <int> readit(cin); // input iterator
/*Below function push_back() is used to add elements at back of a vector
V. The function pushes back the value read by readit. Thus a vector with 5 elements
is created. For more details on constructors of vector class see Chapter 22.*/
V.push_back (*readit++);
V.push_back (*(readit++));
V.push_back (*(readit++));
V.push_back (*(readit++));
V.push_back (*(readit));
cout<<“Elements of V are : ”;
vector<int>:: iterator iter = V.begin();
/* vector is name of class, iterator is key word. iter is the name of iterator.
Function begin()returns iterator to first element of V. vector class supports
random iterator */
while( iter != V.end())
//end() returns iterator just past the last element of vector
cout<< *iter++ <<“ ”; // display values of elements of vector
return 0;
}

The expected output is given below.

Enter 5 integers one after another.
30 40 50 60 70
Elements of V are : 30 40 50 60 70
Below is another program dealing with vectors and uses iterators for traversing the vector.

PROGRAM 21.2 –Example of a vector, input iterator of istream and iterator for traversing
the vector.

#include<iostream>
#include<vector>
 Standard Template Library ❖ 515❖
# include <iterator>
using namespace std;
vector<int> V; // Declaration of a vector with int elements

int main()
{
cout<<“Enter 5 integers one after another.”<<endl;
int count =0;
while (count <=4)
{istream_iterator <int> readit(cin); // input iterator

V.push_back (*readit);
++count;}
/* vector is constructed by pushing back elements. The elements are added
at the back of vector*/
cout<<“\nElements of V are : ”;

vector<int>:: iterator iter = V.begin(); // iterator defined

/*iter is name of iterator. In following it is used for traversing through

the vector*/

while( iter != V.end())

{cout<< *iter <<“ ”;
iter++;}
return 0;
}

The expected output is given below.

Enter 5 integers one after another.
25
30
40
30
60
Elements of V are : 25 30 40 30 60

THE DEQUE
The following program illustrates operations on a deque, i.e. declaration of deque, construction
of deque and input, output iterators of iostream. The deque also supports random iterator and any
element may be accessed directly. For more details and illustrations on deque see Chapter 22.

PROGRAM 21.3 – Illustrates input, output iterators and application to a deque.

#include<iostream>
#include<deque>
# include <iterator>
❖ 516 ❖ Programming with C++
using namespace std;
deque <int> Dek; // declaration of deque of integers
int main()
{
cout<<“Enter 4 integers.”<<endl;
int count =0;
while (count < 4)
{ istream_iterator <int> readit(cin);
Dek.push_back (*readit);
++count; }
int k=0;
cout<<“\nElements of Dek are : ”;
while (k <4)
{ostream_iterator <int> writeit(cout);
* writeit= * (Dek.begin()+k );
cout<<“ ”;
k++;}
return 0;
}

The output of the program is given below.

Enter 4 integers.
45
30
20
60
Elements of Dek are : 45 30 20 60

THE LIST
The following program illustrates the list container. It is like a doubly linked list. The program
illustrates declaration of list, its construction and its output. Below is a sample program. For more
details on lists and functions and operators supported by class list see Chapter 22.

PROGRAM 21.4 – Illustrates declaration and construction of lists and input/output.

#include<iostream>
#include<list>
using namespace std;
list<char> Lc ; //declaration of a list with name Lc and type char
list<int> Li (7,60); // declaration of list with name Li
// it has 7 int elements each of value 60
 Standard Template Library ❖ 517❖
int main()
{
for (int i=0; i<6;i++)
Lc.push_back(65+i); // Adding elements at back of list Lc

cout <<“The first element of Lc is ”<<Lc.front()<<“ and last element is ”

<<Lc.back()<<endl; // output of first and last

list<double>Ld(3,2.5); //List Ld has 3 elements each of value 2.5

for ( int j =0; j< 3; j++)

{Ld.push_back( 5.5); // Add 3 elements each 5.5 at back
Ld.push_front( 10.5);} //Add 3 elements each 10.5 at front

cout<< “Lc = ”;
list<double>::iterator itrd; // iterator for a double list

list<int> :: iterator itri; //declaration of iterator for int

list <char>:: iterator itrc; // iterator for char list

for(itrc= Lc.begin(); itrc!=Lc.end(); itrc++)

cout << *itrc <<“ ”; // output of elements of list Lc

cout<<“\n Li = ”;
for(itri= Li.begin(); itri!=Li.end(); itri++)
cout << *itri <<“ ”; // output of elements of list Li

cout<< “\n Ld = ”;
for(itrd= Ld.begin(); itrd!=Ld.end(); itrd++)
cout << *itrd<<“ ”; // output of elements list Ld.

cout <<“\n First element of Ld = “ <<*(Ld.begin())<<endl;

cout<<“\n Third element of Ld from beginning = ” <<*(++(++ Ld.begin())) ;
// list class does not support random access

cout<<“\n Third element of Ld from end = “ << * (—–(–— Ld.end()))<<endl ;

return 0;
}

The expected output is given below.

The first element of Lc is A and last element is F
Lc = A B C D E F
Li = 60 60 60 60 60 60 60
Ld = 10.5 10.5 10.5 2.5 2.5 2.5 5.5 5.5 5.5
First element of Ld = 10.5

Third element of Ld from beginning = 10.5

Third element of Ld from end = 5.5
❖ 518 ❖ Programming with C++

21.5 ASSOCIATIVE CONTAINERS

The associative containers support fast retrieval of elements through keys. These comprise (i) sets,
(ii) multisets, (iii) maps and (iv) multimaps. The container size may be varied during the program
execution. The elements of an associative set are of type value_type. Each element is associated
with a key of type key_type. In case of sets and multisets the element value itself is the key, i.e.
key and the associated value is one and same while in case of maps and multimaps the keys are
different from the associated values. Thus in sets and multisets an element comprises single value
while in maps and multimaps an element is a pair comprising a key and a value.
The sets have only unique keys while multiset allows duplicate keys. Similarly, map allows only
unique keys while multimaps allow duplicate keys. The following program illustrates the
construction and input/output of a map. For more details on set and maps see Chapter 23. Here
we take an illustrative program as given below on sets.

PROGRAM 21.5 – Illustration of construction and output of a map.

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef map <string, int> Mint ;
/* typedef has been used to avoid writing of map<string, int> several times.
In the declaration key is a string and value int.*/
int main()
{
string Name ;
int Marks;
Mint Grade; // Grade is the name of map

for (int i = 0; i<4; i++)

{cin>> Name ;
cin >> Marks;

Grade [Name] = Marks ;} // making pairs of values

Mint :: iterator iter; // declaration of iterator

for( iter = Grade.begin(); iter != Grade.end(); iter++)

cout<< (*iter).first<<“ \t ” <<(*iter).second<<“\n” ;

/*first is the key and second is the value. The dot operator selects first
and second.*/
return 0;
}

The output of the program is given below. The first eight lines are data input and remaining four
lines are output of program.
Sunita
85
 Standard Template Library ❖ 519❖
Mamta
75
Adity
90
Pummy
70
Adity 90
Mamta 75
Pummy 70
Sunita 85

21.6 CONTAINER ADAPTERS: STACK, QUEUE AND PRIORITY QUEUE

The stack, queue and priority queue are the container adapters in STL. They do not support any
iterator and do not have their own data implementation structure like vectors, lists, etc. However,
they can adapt the data structures of other containers. Therefore, the programmer can choose the
container that suits the program at hand.

STACK
Stacks allow the insertion and deletion of data at one end, i.e. last in first out (LIFO) or first
in last out order (FILO). It is illustrated in Fig. 21.3 below. The figure shows that plates are
stacked one over another. For taking out a plate it is picked from the top of the stack. So the
plate that was placed last is picked up first. There are many applications of stacks, for instance,
compiler makes a stack of machine code from the source code. The functions supported by stack
class are listed in Table 21.3.

Fig. 21.3: Last in first out LIFO

Table 21.3 – Functions associated with stacks
Function Description
empty() The function returns true if stack is empty.
pop() The function removes the top element of stack.
push() The function adds an element at the top of stack.
size() The function returns the number of elements in the stack.
top() The function returns the top element of stack.

The following program illustrates implementation of a stack. The function push () is used
to fill the stack. The applications of some of the above mentioned functions are illustrated with
a stack thus formed.
❖ 520 ❖ Programming with C++
PROGRAM 21.6 – Illustrates application of different functions on a stack.

#include<iostream>
#include<stack>
using namespace std;

void main()
{ stack<int> St1 ; //St1 is declared a stack of integers

for ( int i =1; i< 7; i++) // construction by push function

St1.push (10* i);
//The elements pushed are 10, 20, 30, 40, 50, 60
// In stack it is last in first out order

cout <<“Top element of St1 = ” <<St1.top() <<endl;

cout<<“Size of stack St1 is = ” << St1.size() <<endl;

cout<<“Elements of St1 are: ”;

while(!St1.empty()) // output statement for stack elements

{cout<<St1.top()<<“ ” ;
St1.pop() ; } // remove the top element
}

The expected output is given below.

Top element of St1 = 60
Size of stack St1 is = 6
Elements of St1 are: 60 50 40 30 20 10

In the above program the stack is constructed by pushing values. In the output statement
the top element is put to output stream first, then, the top element is removed by the function
pop(). The next element comes to top and is put to output stream and then popped out
(removed). The process continues till the last element is out. From the output it is clear that the
first element in output stream is the one that was pushed in last. The following program presents
another example of stack.

PROGRAM 21.7– Illustrates another example of stacks of elements of different types.

#include<iostream>
#include<stack>
#include <string>
using namespace std;
void main()
{
stack<int> Ski ; // declaration of stack of integers
stack<char> Sch; // declaration of stack of char elements
stack<string> Skt ; // declaration of stack of strings
 Standard Template Library ❖ 521❖
string Name[4]={“Delhi”, “Mumbai”, “Kanpur”, “Madras”};
char ch[4] = { ‘B’,’A’, ‘C’, ‘T’};
int Array[5] = { 10, 40, 60,20,10 };

for ( int i = 0;i < 4; i++)

{Ski.push (Array[i]); // push 4 elements in stack Ski
Sch.push (ch[i]); // push 4 elements in stack Sch
Skt.push (Name[i]);} // push 4 elements in stack Skt

cout <<“Top element of Skt = ” <<Skt.top()<<endl;

cout <<“Top element of Sch = ” <<Sch.top()<<endl;

cout <<“Top element of Ski = ” <<Ski.top()<<endl;

cout<<“Size of stack Skt is = ” << Skt.size() <<endl;

cout<<“The elements of three stack are as below.”<<endl;

for ( int j = 0;j < 4; j++)

{ cout<<Skt.top()<<“\t” << Sch.top()<<“\t” <<Ski.top() <<“\n”;

Skt.pop() ; // remove the top element of Skt

Sch.pop(); // remove the top element of Sch
Ski.pop(); // remove the top element of Ski
}
}

The expected output is as under.

Top element of Skt = Madras
Top element of Sch = T
Top element of Ski = 20
Size of stack Skt is = 4
The elements of three stack are as below.
Madras T 20
Kanpur C 60
Mumbai A 40
Delhi B 10

PROGRAM 21.8 – Illustrates stacks with underlying data structure of vector.

#include<iostream>
#include<stack>
using namespace std;
#include <vector>

int main()
{
stack <int, vector <int> > Stack1, Stack2, Stack3;
// Declaration of Stacks of int
// The underlying container is vector
❖ 522 ❖ Programming with C++
int Array1[]={12 ,13 ,14 ,15 , 16 };
int Array2[] = { 40, 60,20,10,50 };

for ( int i = 0;i < 5; i++)

{Stack1.push (Array1[i]);
Stack2 .push (Array2[i]); }

Stack3 = Stack2; // Stack2 is assigned to Stack3

cout<<“Size of Stack3 is = ” << Stack3 .size() <<endl;

if ( Stack2 == Stack3 ) // relational == operator

cout<< “Stack3 and Stack2 are equal \n”;
else
cout<<“Stack3 and Stack2 are not equal \n”;

Stack3.push(80); // add another element into stack

cout << “Now size of Stack3 = ”<<Stack3.size()<<endl;

for ( int j = 0;j < 6; j++) // for output of stack elements

{cout<<Stack3.top()<<“ ” ; //display the top element of Stack3

Stack3.pop();} // remove the top element of Stack3

return 0;
}

The expected output is given below.

Size of Stack3 is = 5
Stack3 and Stack2 are equal
Now size of Stack3 = 6
80 50 10 20 60 40

QUEUE
In queue the data elements can be inserted at the back and removed from the front-end. The
order is same as it happens in real life queues, i.e. first in first out (FIFO). It is illustrated in the
following figure. Queue may be formed by pushing items by function push().

back() front()
push()

pop()

Fig. 21.4: An illustration of queue

 Standard Template Library ❖ 523❖
The various functions associated with a queue are listed in Table 21.4
Table 21.4 – Functions associated with queues. Header file <queue>

Function Description
back() The function returns a reference to last element of queue.
empty() The function returns true if queue is empty.
front() The function returns a reference to the first element of current queue.
pop() The function removes the front element of queue.
push () The function adds an element at the end of queue.
size () The function returns the number of elements in the queue.

The following program illustrates different operations on queues.

PROGRAM 21.9 – Illustrates application of some queue functions.

#include<iostream>
#include<queue>
using namespace std;

void main()
{
queue <int> Q ;

for ( int i =1; i<7; i++)

Q.push (10* i);
// in queue there is no function top() instead it is front()
// the output of elements from the front of queue is
//illustrated below.
cout<< “The element at the front = ” <<Q.front()<<endl;
cout<<“Size of queue Q is = ” << Q.size() <<endl;
cout<<“Elements of Q are: ”;

while(!Q.empty()) // output statement for queue

{cout<<Q.front()<<“ ” ;
Q.pop() ; }

The expected output is given below.

The element at the front = 10
Size of queue Q is = 6
Elements of Q are: 10 20 30 40 50 60

The following program illustrates some of operators supported by queues.

❖ 524 ❖ Programming with C++
PROGRAM 21.10 – Illustrates operators acting on queues.

#include<iostream>
#include<queue>
using namespace std;
int main()
{
queue<int> q2,q3 ;
queue<double> q1;
for ( int i =0; i< 6; i++)
{q1.push (1.5*i);
q2.push (i*i);}
q3 = q2; // assignment operator
if( q2 == q3)
cout<< “Now the queues q2 and q3 are equal”<<endl;
else
cout<< “Now the queues q2 and q3 are not equal”<<endl;
cout<< “The elements of q2 are: ”;
while(!q2.empty())
{cout<<q2.front()<<“ ” ;
q2.pop() ;}
cout<<“\n”;
cout<< “The elements of q3 are: ”;
while(!q3.empty())
{cout<<q3.front()<<“ ” ;
q3.pop() ;}
cout<<“\n”;
if( q2 == q3)
cout<< “The queues q2 and q3 are equal”<<endl;
else
cout<< “The queues q2 and q3 are not equal”<<endl;
cout<<“push another element 10 into q3 \n”;
q3.push (10);
cout<< “The elements of queue q3 are: ”;
while(!q3.empty())
{cout<<q3.front()<<“ ” ;
q3.pop() ;}
cout <<“\n”;
 Standard Template Library ❖ 525❖
cout<< “The elements of q1 are: ”;
while(!q1.empty())
{cout<<q1.front()<<“ ” ;
q1.pop() ;}
cout <<“\n”;
return 0;
}

The expected output is given below.

Now the queues q2 and q3 are equal
The elements of q2 are: 0 1 4 9 16 25
The elements of q3 are: 0 1 4 9 16 25
The queues q2 and q3 are equal
push another element 10 into q3
The elements of queue q3 are: 10
The elements of q1 are: 0 1.5 3 4.5 6 7.5
In the above program three queues are declared—two queues of int and one of double
numbers. The queues q1 and q2 are constructed by push function while q3 has been made as a
copy of queue q2. The elements are added at back while they are popped out (removed) at the
front-end. The application of functions empty () and front () and operator ++ is illustrated. After
the assignment q3 = q2 ; the two queues are equal. After the elements have been popped
(removed) then also they are equal because there is no element left in both the queues. Then an
element (10) is added to q3. Now the element in q3 is only 10. In the last line of output the
elements of q1 are displayed. After the display there is now no element left in q1 because all are
removed (popped out) during display.

PRIORITY_QUEUE
The priority queues are queues in which the elements are ordered by a predicate. By default the
element with highest value is at the top. The various functions associated with priority queues
are described in Table 21.5. Here the function top() is used to point to the element at the front-
end in place of function front() which is used in queue.
Table 21.5 – Functions associated with priority_queue

Function Description
empty() The function returns true if priority_queue is empty.
pop() The function removes the top element of priority_queue.
push() The function adds an element into priority_queue.
size() The function returns the number of elements in priority_queue.
top() The function returns top element of the priority_queue.

The following program illustrates the application of some of these functions.

❖ 526 ❖ Programming with C++
PROGRAM 21.11 – Illustrates priority queues.

#include<iostream>
#include<queue>
using namespace std;
void main()
{
priority_queue <int> PQ ; // declaration of priority queue
int Array[6] = { 10, 80, 90, 20, 40,70};
for ( int i =0; i<6; i++)
PQ.push (Array[i]);
// in priority _queue there is no front instead it is top()
cout<< “The element at the top = ” <<PQ.top()<<endl;
cout<<“Size of queue PQ is = ” << PQ.size() <<endl;
cout<<“Elements of PQ are: ”;
while(!PQ.empty())
{cout<<PQ.top()<<“ ” ;
PQ.pop() ; }
}

The expected output is given below.

The element at the top = 90
Size of queue PQ is = 6
Elements of PQ are: 90 80 70 40 20 10
The output is self explanatory. The unordered elements of an array are pushed into the
priority queue. We see from the output that the items are ordered with the element having
highest value at the top.

21.7 FUNCTION OBJECTS/PREDICATES

A function object is a class object which acts as a function and returns a bool value depending
upon whether its argument, which may be a container element, satisfies the function or not. In
several algorithms the elements of a container are selected subject to whether the element satisfies
a predicate which is a function object. The algorithm is implemented subject to the return value
of function object. If the return value is true the algorithm is implemented otherwise not. Before
going to function objects, we shall examine an ordinary function acting as a predicate.

PROGRAM 21.12 – Illustrates a global function used as predicate in an algorithm.

#include<iostream>
#include<algorithm>
#include<functional>
 Standard Template Library ❖ 527❖
using namespace std;

bool Odd (int n) // definition of function for odd number

{return n%2 ? true : false;}

bool Even(int m ) // definition of function for even number

{return !(m%2) ? true: false ;}

int main()
{
int S[ ] = { 5,6,8,7,4,3,8,10,11, 12};
int n = count_if(S, S+10, Even );
/* count_if()is an algorithm. Even is the address of (pointer to) the function
Even() defined above. The function count_if is implemented if the element
satisfies the function Even*/
cout<< “Number of even elements of S are = ” <<n <<endl;

int K = count_if(S, S+10, Odd);

/* Odd is the address of function Odd defined above. The function count_if()
is implemented if the element satisfies the function Odd*/
cout<<“Number of odd elements of S are = ” <<K<<endl;
return 0;
}

The expected output is given below.

Number of even elements of S are = 6
Number of odd elements of S are = 4
In the above case the elements are examined if they are divisible by 2 or not and according
to the bool return value which is either true or false the element is counted or rejected. One of
the functions (Odd) tests the elements for odd numbers while the second function (Even) tests
the elements whether they are even. Though only one function is needed, the two are included
to illustrate the form of the functions. The selection of even or odd depends on the return values
of these functions. Similarly, one may define a function whether a particular integer is divisible
by another integer or not. In such a case, if the divisor is required to be changed often, then for
every new divisor the program has to be changed. It is better to define a class with the function
as an object of the class. It is convenient because class object can hold the data (the divisor). In
such a scheme the function is called function object. It is illustrated below.

PROGRAM 21.13 – Illustrates definition of a predicate as a template class.

#include<iostream>
# include<algorithm>
using namespace std;
template <class T>
class GTE { // GTE = Greater than or equal to
private :
T x;
❖ 528 ❖ Programming with C++
public :
GTE (T A){ x = A;} // constructor function
bool operator() (T y)
{return y >= x ? true :false;}
}; // end of class

int main()
{ GTE <int> gte(30); //gte is declared an object of GTE for int

GTE <double> gted (25.0); // gted is an object for double 25.0

int S[] = { 10,20,30 ,36, 44, 60, 70 };

double SD [] = { 3.5, 27.5, 22.6, 56.7, 80.0, 90.7, 65.5, 35.5};
//count_if() is a an algorithm

int m = count_if ( S, S+7, gte); // here gte is a predicate

cout << “Number of elements in S >= 30 are = ”<<m<<endl;

int n = count_if( SD, SD+8, gted); //here gted is a predicate

cout<< “Number of elements in SD >= 25.0 are = ”<< n<<endl;

return 0 ;
}

The expected output is given below.

Number of elements in S >= 30 are = 5
Number of elements in SD >= 25.0 are = 6
In the above program the objects gte and gted behave like functions. The initialisation values
of the objects, i.e. 30 and 25.0 become the arguments of the functions. It seems a better option
than the predicate function of the previous program.
There are several predefined predicates in C++ as listed in Table 21.6 below. As illustrated
above the user may as well develop his/her own predicates.

21.8 PREDEFINED PREDICATES IN C++

There are nine predefined relational and logical predicates defined in header file <functional>.
Some of the predicates provided in the header file are listed in Table 21.6.
Table 21.6 – Predefined predicates in header file <functional>

Predicate Arity ( Unary/binary) Equivalent operation

plus binary x+y
minus binary x–y
multiplies binary x*y
divides binary x/y
modulus binary x%y
Contd…
 Standard Template Library ❖ 529❖

negate unary –x
equal_to binary x==y
not_equal_to binary x!=y
greater binary x>y
greater_equal binary x>=y
less binary x<y
less_equal binary x<=y
logical_and binary x&&y
logical_or binary x||y
logical_not unary !x
An illustration of use of some predicates is provided by Program 21.14 given below.

PROGRAM 21.14 – Illustrates application of predicates less and greater with algorithm
sort().

#include<iostream>
#include<algorithm>
#include<functional> // header file for predefined predicates

using namespace std;

int main()
{
int S[8] = { 5,6,8,7,4,3,8,9};
char ch [] = { ‘A’, ‘Z’, ‘C’, ‘M’, ‘G’, ‘K’, ‘T’};

sort(S, S+8, less<int>()); // arrange in increasing order

sort( ch, ch+7, greater<int>()); // arrange in decreasing order
for ( int i =0; i<8;i++)
cout<<S[i]<<“ ”;
cout <<endl;

for ( int j =0; j< 7 ; j++)

cout << ch[j]<<“ ”;
cout<<endl;

return 0;
}

The expected output is given below.

34567889
ZTMKGCA

In the first line of output the array element are sorted in increasing order. In the second line
the characters are sorted in decreasing order.
❖ 530 ❖ Programming with C++

21.9 BINDER FUNCTIONS

When every element of a sequence is to be compared with a value for further action it is better
to bind the value with which it is to be compared with the predicate by a binder function. Two
functions are defined as given below. Let us have a comparison of type expression x >= y.
bind1st (x)// calls or binds the function with x as first argument.
bind2nd(y)// calls or binds the function with y as second argument.

The application of bind2nd() is illustrated in the following program.

PROGRAM 21.15 – Illustrates application of bind2nd().

#include<iostream>
#include<algorithm>
#include<functional>
using namespace std;

template <class T >

struct Greater: binary_function< T , T , bool > {
bool operator ()(const T & x, const T& y) const
{ return ( x >y );}};

int main()
{
int S[10] = { 5,6,8,7,4,3,8,9,12, 14 };

int n = count_if ( S, S +10, bind2nd(Greater<int>(), 7) );

cout<<“The number of elements greater than 7 are : ”<< n <<endl;

return 0;
}

The expected output is given below. Every element of array is compared if it is more than 7.
The number of elements greater than 7 are : 5
The following program uses the standard function objects greater than and less than.

PROGRAM 21.16 –Illustrates application of predicates greater, less and binder

function.

#include<iostream>
#include<algorithm>
using namespace std;
#include<functional>
int main()
{
 Standard Template Library ❖ 531❖
int S[ ] = { 5,6,8,7,8,3,8,10,8, 12};

int n = count_if(S, S+10, bind2nd(greater<int>(),7));

cout<< “Number of elements of S >7 are = ” <<n <<endl;

int m = count_if(S, S+10, bind2nd(less<int>(),10));

cout<< “Number of elements of S < 10 are = ” <<m <<endl;
return 0;
}

The expected output is as under.

Number of elements of S >7 are = 6
Number of elements of S < 10 are = 8

EXERCISES
1. What is STL all about?
2. What are the main components of STL?
3. What is an iterator?
4. Which of the containers are first class containers and why?
5. How are associative containers different from sequence containers?
6. How are iterators declared? Give examples of declaration of iterator for following cases.
(i) vectors with int elements.
(ii) deque with double elements.
(iii) list of a names.
7. Do the stacks support iterators?
8. What is the difference between a queue and a priority queue?
9. Make a program to illustrate the application of following functions.
(i) push()
(ii) top()
(iii) size()
10. Make a program to illustrate the application of following functions.
(i) size()
(ii) push()
(iii) pop()
11. How is a stack different from a vector?
12. What is a predicate?
13. Define a function predicate for sorting an array in increasing order.
14. What is a binder function?
15. Make a test program to sort out even numbers in a sequence of integers.

❍❍❍
❖ 532 ❖ Programming with C++

CHAPTER

22.1 INTRODUCTION

Some of the features of containers have already been discussed in the Chapter 21 on STL.
The sequence containers comprise three containers which are vector, list and deque (double
ended queue). The comparative features of the three are described in the Table 22.1 below.
Table 22.1 – Salient features of vector, list and deque containers.

vector list deque

Header file <vector> Header file <list> Header file <deque>
Supports random iterator Supports bidirectional Supports random iterator.
Rapid access to any element. iterator. The iterator can Rapid access to any element
be incremented or decremented. in deque.
Because of random iterator, The STL algorithms that Because of random iterator,
all STL algorithms can require input/output, all STL algorithms can
operate on vectors. forward and bidirectional operate on deques
iterators can operate on lists
Rapid insertions and Rapid insertions/deletions Rapid insertions and
deletions at the back-end anywhere in the list. deletions at the both ends
of vector. of deque.
Data stored in sequential Doubly linked list like data Based on array like data
memory locations like structure. structure. Direct access
an array. possible with index operator.
The commonly used The commonly used functions The commonly used funct-
functions are described are given in Table 22.3 ions are described in Table
in Table 22.2. and Table 22.4. 22.5.

22.2 VECTOR CLASS

DECLARATION OF VECTORS
As mentioned above, the vector class is a container class and vector elements are stored in
sequential memory locations like an array. A C++ vector is not the mathematical vector though
 Sequence Containers – vector, list and deque ❖ 533❖
some of the concepts are similar. The C++ vector can have any number of components. It
may be taken as a more robust and dynamic array. Besides, the vector class supports many
functions which cannot be directly applied to pointer based arrays such as insertion and deletion
of elements. The vector is a dynamic array in the sense that it can allocate to itself additional
memory if needed due to the addition of elements. To start with, the allocation is a small
chunk of memory which is sufficient for a few elements. If the elements increase during
execution of program, the vector can relocate itself with a bigger segment of memory. While
an array is a fixed length sequence (number of elements is fixed at the time of declaration).
In case of arrays, an individual element can be assigned, while the array as a whole cannot be
assigned. A vector may be assigned as a whole. For working with vectors we have to include
the header file <vector> and have to declare the type of elements of the vector, i.e. int, double,
char, etc., in angular brackets (< >). Examples of vector declarations are given below.
#include <vector> // Header file
vector <int> V1; // V1 is vector with int elements
vector <double> V2; // V2 has elements of type double
vector <char> Vect; // Vect has elements of type char
In the above declarations V1, V2 and Vect are the identifiers or names of the vectors.
The elements of vector V1 are of type int, elements of V2 are of type double and elements
of Vect are of type char.
More than one vector may be declared in same line if their type is same, we need to
write type in angular brackets only once after the class name vector. For instance, we may
declare multiple vectors as below.
vector <int> V1, V3 ;
vector <double> A, B, C ;

CONSTRUCTORS AND DESTRUCTORS OF VECTOR CLASS

The container class vector supports a number of constructors. Some methods of constructing
vectors are illustrated below.
vector <double> V(5); /* Constructs a vector with 5 elements of type double
and value 0*/
vector <int> V1(4,35); /* Constructs V1 with 4 elements of type int and
value 35*/
vector <int> V4 (V1); // Constructs vector V4 as a copy of vector V1
vector <char> V2 ( 4, ‘D’) ;
/* Constructs vector V2 with 4 elements each of type
char and value ‘D’*/
~vector (); /* Destructor of the vector—it deletes all the components
of the vector.*/
The following program illustrates some methods of constructing vectors. The output of vector
elements has been carried out by [ ] operator as it is used for array elements.
❖ 534 ❖ Programming with C++
PROGRAM 22.1 – Illustrates construction of vectors and output of vector elements.

#include<iostream>
#include<vector>
using namespace std;
vector <double> V(4) ; // size =4, value of each = 0
vector <int> V1(4,35) ; // size =4, value of each =35
int main()
{vector<double> V2 (4, 4.5); //size = 4, value of each = 4.5
vector<char> V3 (4, ‘B’);
// vector V3 has four elements each equal to ‘B’.
vector <int> V4 (V1); // V4 is copy of V1
vector<int> V5 ;
for(int i =0; i<4;i++ ) // It constructs a vector with 4
V5.push_back (5); //elements. Each element is equal to 5.
cout<<“V \tV1\tV2\tV3 \tV4”<<“\tV5”<<endl;
for ( int j = 0; j< 4; j++) // for loop for output
cout <<V[j]<<“\t”<< V1[j]<<“\t” <<V2[j]<<“\t”<< V3[j] <<“\t”<<
V4[j]<<“\t”<<V5[j]<<endl ;
return 0;
}

The expected output is as under. The elements of vectors are printed from top to bottom.
V V1 V2 V3 V4 V5
0 35 4.5 B 35 5
0 35 4.5 B 35 5
0 35 4.5 B 35 5
0 35 4.5 B 35 5

The output is self explanatory. The vector V has 4 elements each having a value 0, the
vectors V1, V2 and V3 are similarly constructed. The values of their elements are 35, 4.5
and ‘B’ respectively. The vector V4 is a copy of V1. The vector V5 is constructed by the
function push_back (). The function adds elements at the back. In this case, all the values
pushed are 5.

22.3 FUNCTIONS OF VECTOR CLASS

There are a number of member functions of vector class which may be used for manipulation
of elements in the sequence. Some of these are given below in Table 22.2.
 Sequence Containers – vector, list and deque ❖ 535❖
Table 22.2 – Functions of class vector

S.No. Function Description

1. assign() The function assigns values to a vector from beginning to end. Function
is also used to assign a number of copies of a value to a vector. The
previous contents are removed.
2. at() This function returns a reference to an element in a vector at the
specified location (written in brackets).
3. back() It returns reference to the last element of the vector.
4. begin() It returns iterator to first element of the vector.
5. capacity() It returns number of elements that a vector can hold in the current
allocation of memory. It is not the actual number of elements of the
vector.
6. clear() The function deletes all the elements of a vector.
7. empty() The function returns true if there is no element in the vector, false
otherwise.
8. end() It returns iterator just past the end of vector.
9. erase() It deletes element at the indexed location or elements between start
and end iterators. The start element is included but the end element
is not included.
10. front() It returns reference to first element of the vector.
11. insert() It inserts elements in vector (i) one value or (ii) a number of copies of
a value at a specified location, (iii) insert from start to end before a
location.
12. max_size() Returns maximum number of elements that a vector can hold.
13. pop_back() Removes the last element of the vector.
14. push_back() Appends an element at the back of vector.
15. rbegin() Returns a reverse iterator to the end of the vector.
16. rend() Returns a reverse iterator to the beginning of vector.
17. reserve() Sets the capacity of vector to at the least size.
18. resize() Changes the size of vector to the size specified.
19. size() Returns the current number of elements in the vector.
20. swap() Exchanges the elements of two vectors.

22.4 DEFINITION AND APPLICATION OF ITERATORS

Iterators are like pointers. These are used to traverse through the elements of a vector. The
declaration of iterator say for vector V1 of type int is illustrated below. In the following
declaration it is also initialized to V1.begin(), i.e. the value corresponding to the first
element of vector V1.
❖ 536 ❖ Programming with C++

Scope resolution operator vector V1 Selection dot operator

vector <int> ::iterator Iter = V1.begin();

template name type of V1 keyword name of iterator function

Fig. 22.1: Declaration and assignment of iterator

According to the above declaration, Iter is the name of iterator. It is initialized with value
corresponding to the first element of the vector. The value may be incremented to reach the
other elements of the vector. Thus Iter+1 will point to the second element of the vector.
Similarly Iter+2 points to the third elements and so on. The V1.end() points to just past the
last element of the vector. The values of respective elements may be obtained by dereference
operator (*). Thus the value of first element is equal to *Iter , second element is equal to
*(Iter+1), etc. See the following program for illustration.

PROGRAM 22.2 – Illustrates iterators and functions push_back(), begin() and end().

#include<iostream>
#include<vector>
using namespace std;
vector<int> V1 , V2 ; // declaration of two vectors
int main()
{
V1.push_back(10); // Use of function push back to construct V1
V1.push_back(20);
V1.push_back(30);
vector <int> :: iterator iter = V1.begin(); // iterator for V1
cout<< *iter <<“\t” << *(iter+1)<<“\t”<< *(iter +2) <<endl;
cout<< iter<<“\t”<< iter+1<<“\t” << iter+2 << endl;
cout << “Elements of vector V2 are as below.\n “;
for (int i =0;i< 6; i++)
V2.push_back (i*i); // construction of V2
iter = V2.begin(); // iterator for V2
while ( iter != V2.end())
{cout <<*iter <<“ ”;
iter++;}
cout<<endl;
return 0;
}
 Sequence Containers – vector, list and deque ❖ 537❖
The expected output is as under.
10 20 30
00480070 00480074 00480078
Elements of vector V2 are as below.
0 1 4 9 16 25
With the help of function push_back () three elements are pushed into V1. These are
additions to the existing elements. The values of these elements are 10, 20 and 30 respectively.
The output of these has been carried out by iterator iter which is initialized to V1.begin().
Thus *iter is the value of 1st element, *(iter+1) gives value of 2nd element and
*(iter+2) gives value of third element and so on. The values of iter, iter+1 and iter+2 are
the addresses of the elements. Similarly the elements of V2 are extracted with the help of
while loop along with iterator iter.

22.5 OPERATORS SUPPORTED BY VECTOR CLASS

Two vectors may be compared with overloaded comparison operators such as ==, >,
<, <=, >=,!=. A vector may be assigned to another just like we assign the variables of
fundamental type. This is illustrated below.
vector<int> V1 (5, 50) , V2 ;
V2 = V1;
In the above code two vectors are declared. The V1 has been initialized with 5 elements
each equal to 50. The vector V1 is assigned to V2, so V2 becomes a copy of vector V1. The
equality may be tested with the operator (==). Two vectors are equal if their sizes are equal
and each element of vector one at location i is equal to the element at location i of the second
vector. The arithmetic operations are possible on individual elements of a vector. The following
program illustrates the application of some of the operators.

22.6 APPLICATION OF FUNCTIONS OF VECTOR CLASS

The different functions supported by vector class are listed in Table 22.2 along with the brief
description about the action they can perform. Some of these are discussed below.

FUNCTION ASSIGN ()
Let V be a vector to which the values are to be assigned. If all the components are equal we
may assign the values as below.
V.assign (6, 80);
According to the above code the vector V has been assigned 6 elements and each element
has value 80. Thus the vector V has the elements 80, 80, 80, 80, 80 and 80. The function
assign () removes the existing elements of the vector. The function push_back() is also
used to put in values of vector elements. It appends the value written in the brackets at the
❖ 538 ❖ Programming with C++
back of the vector, i.e. it adds a new element at the back of the vector. Thus if it is intended
to add a new element to V at its back, the syntax is given below.
V.push_back (35);
With the addition of 35, the vector V now has the elements 80, 80, 80, 80, 80, 80
and 35.

PROGRAM 22.3 – Illustrates functions assign(), push_back() and some operators.

#include<iostream>
#include<vector>
# include <string>
using namespace std;
vector<int> V1, V2,V3 , V4;
vector <string> V5;
int main()
{
V1.assign(5,5 ); //V1 is assigned 5 values each equal to 5.
V2.assign(5,4); //V2 is assigned 5 values each equal to 4.
V3 = V2; // vector V2 is assigned to vector V3
string Str [] = {“Nimi” ,”Simi”, “Jimmy”, “Ginny”, “Pummy”};
int Array [] = { 1,2,3,4,5};
for (int j =0; j<5 ;j++)
{V4.push_back (Array[j]);
V5.push_back (Str[j]);}
//elements of array Str are pushed in V5
if(V2 == V3)
{cout <<“V2 and V3 are equal”<<endl;}
else
cout<<“ V2 and V3 are not equal”<<endl;
cout<<“V1 \tV2 \tV3 \tV4 \t V5”<<endl ;
for ( int i = 0; i< 5; i++)
cout<<V1[i] <<“\t”<<V2[i] <<“\t”<<V3[i]<< “\t”<< V4[i]<<“\t” << V5[i]
<<endl;
cout<< “Address of V3 = ”<< &V3 <<endl ;
for ( int k=0;k< 5 ;k++)
V2[k] *= V1[k] ;
cout<< “Components of V2 now are = ”;
for (int m =0; m<5; m++)
cout<<V2.at(m) <<“ “;
 Sequence Containers – vector, list and deque ❖ 539❖
cout<<“\n”;
return 0;
}

The expected output of the program is given below.

V2 and V3 are equal
V1 V2 V3 V4 V5
5 4 4 1 Nimi
5 4 4 2 Simi
5 4 4 3 Jimmy
5 4 4 4 Ginny
5 4 4 5 Pummy
Address of V3 = 0047B780
Components of V2 now are = 20 20 20 20 20

In the above program we get an illustration of application of function assign() and

the function at(). Also the program has a vector whose elements are strings. The output of
vectors has been done by [] operator like an array rather than by iterators. The individual
elements of a vector may be subjected to arithmetic operations like any other variable.

FUNCTIONS SIZE () AND FUNCTION CAPACITY ()

The function size() gives the actual number of elements in the vector while capacity gives
the maximum number of elements it can hold in the current memory allocation. Vector is a
dynamic array, as the number of elements increases beyond the allocated capacity, the vector
is relocated with a bigger chunk of memory.

PROGRAM 22.4 – Illustrates iterators and functions size(), capacity(),

end(),begin(), push_back() and pop_back().

#include<iostream>
#include<vector>
using namespace std;
vector<int> V ;
int main()
{
for ( int i =1; i<=5;i++)
V.push_back (i*i); // 5 values 1,4,9,16,25 are pushed into V
cout<<“Size of V = ”<< V.size()<<endl;
cout<<“Capacity of V = ” << V.capacity()<< endl;
cout<<“The elements of vector V are, ”<<endl;
for (int j = 0; j<=4; j++)
❖ 540 ❖ Programming with C++
cout <<“ “<< V[j]<<“ “ ;
cout<<endl;
V.pop_back(); //The last element (25) of V is removed
cout<<“After pop_back the size of V is = “<< V.size()<<endl;
cout<< “ Capacity of V = “ << V.capacity()<<endl;
for ( int k =1; k< 6; k++)
V.push_back (2*k); // 5 values (2,4,6,8,10)are pushed in to V
cout << “Now size of V is = “<<V.size();
cout<<“\nNow capacity of V is = “<< V.capacity();
vector <int> :: iterator itr; // declaration of iterator itr
cout<<“\nAfter pop_back the elements of V are,”<<endl;
for ( itr = V.begin() ; itr < V.end () ; itr ++)
cout<<*itr <<“ “ ; // output of values of elements with iterators
cout<<endl;
return 0;
}

The expected output is given below.

Size of V = 5
Capacity of V = 8
The elements of vector V are,
1 4 9 16 25
After pop_back the size of V is = 4
Capacity of V = 8
Now size of V is = 9
Now capacity of V is = 16
After pop_back the elements of V are,
1 4 9 16 2 4 6 8 10

The program output given above illustrates the difference between function capacity()
and the function size(). Capacity gives the maximum number of elements that the vector
can hold in the current memory allocation while size gives the actual number of elements.
Vectors are dynamic arrays. The number of elements may increase or decrease during the
execution of the program. If the number of elements increases and the vector needs more
than the memory allocated initially, the vector is relocated with bigger chunk of memory.
The relocation is done automatically and the existing elements are copied to the new location.
In the above output it shows that when the elements increase to 9 the capacity increases from
8 to 16.
The function push_back () adds the element after the end of the vector and pop_back()
removes the last element from vector.
 Sequence Containers – vector, list and deque ❖ 541❖
FUNCTION AT()

In case of arrays, there is no check to see that an index value does not exceed the number of
elements of the array. It is quite possible that the index value may exceed the number of
elements in the array and some garbage value may be substituted as its value. In vector class
a guard is provided by the function at(). It is coded as V.at(k) where k is the index value
and the function gives value of element of vector V at location k. If the value of index exceeds
the number of elements it will give a warning and abort the program as illustrated in the
following program.

PROGRAM 22.5 – Illustrates application of function at().

#include<iostream>
#include<vector>
using namespace std;
vector<int> V ;
int main()
{
V.assign(6,8 ); //assigns 6 values each equal to 8 to V
for ( int i = 0; i< 6; i++)
cout<<V[i] <<“ “;
cout<<“\n\n”;
for( int j=0;j< 6 ;j++)
{V[j] = j*V[j]; //each element is multiplied by its index value
cout <<V[j]<<“ “ ;
}
cout<<“\n\n”;
for (int k =0; k<8; k++)
//Note that the maximum value of k is more than the number of elements
cout<<“The “ <<k<< “ element = “ << V.at(k)<< endl;
return 0;
}

In the above program the vector V has 6 elements, each has value 8 and this is displayed
in the first line of the output given below. The values are then multiplied by its position
index value in the vector, so values are now 0, 8, 16, 24, 32 and 40 as displayed in the
second line of the output. These value are programmed to be displayed with the function
at(k), where k varies from 0 to 7, i.e. 2 more than the number of elements. The program
runs to display the first 6 elements after which it terminates the program because the index
value is more than the number of elements in the vector. The complete output is given below.
8 8 8 8 8 8
0 8 16 24 32 40
❖ 542 ❖ Programming with C++
The 0 element = 0
The 1 element = 8
The 2 element = 16
The 3 element = 24
The 4 element = 32
The 5 element = 40
abnormal program termination

PROGRAM 22.6 – Illustrates application of erase(), and front().

#include<iostream>
#include<vector>
using namespace std;
vector<int> V2(2); // V2 has two elements each of value 0.
vector< double > V4;
int main()
{
for ( int i = 1; i<6; i++)
{ V2.push_back (i*i); // 5 values( 1,4,9,16,25)are pushed in V2
V4.push_back (i*2.5);} //5 values are pushed in V4
cout<<“\n V4 = ”;
vector<double>:: iterator itr4;
for( itr4 = V4.begin(); itr4 != V4.end(); itr4++)
cout<< *itr4<<“ ”; //output of elements of V4
vector<int> :: iterator itr;
cout<<“\nNow V2 = ”;
for( itr = V2.begin(); itr != V2.end(); itr++)
cout<<*itr<<“ ” ;
cout<<“\nAfter erasing first two elements V2 = “;
if ( V2.front() ==0)
V2.erase(V2.begin(),V2.begin()+2);
// erase 1st and 2nd element. The third element is not erased
for( itr = V2.begin(); itr != V2.end(); itr++)
cout<< *itr<<“ ”; // output statement
V2.pop_back( );
V2.pop_back();
cout<<“\nAfter two pop-backs V2 = ”;
for( itr = V2.begin(); itr != V2.end(); itr++)
 Sequence Containers – vector, list and deque ❖ 543❖
cout<< *itr<<“ ”;
cout<<endl;
return 0;
}

The expected output is given below. V2 had 2 elements each equal to 0, later 5 more elements
are pushed into it.
V4 = 2.5 5 7.5 10 12.5
Now V2 = 0 0 1 4 9 16 25
After erasing first two elements V2 = 1 4 9 16 25
After two pop-backs V2 = 1 4 9

In the following program we carry out various operations with relational operators such
as ==, <, >, >=, <=, etc., in the same way as we use these operators for fundamental types.

PROGRAM 22.7 – Illustrates use of relational operators with vectors.

#include<iostream>
#include<vector>
using namespace std;
vector<int> V1;
vector <int> V2;
vector <int> V3;
vector<int> V4;
int main()
{V2.assign (5, 65); //assigns 5 elements each equal to 65 to V2.
V3.assign(5,70); //assigns 5 elements each equal to 70 to V3.
V1.push_back(2);
V1.push_back(4); // push_back appends value 4 at the back of
//the vector. A method of creating a vector.
V1.push_back(5);
V1.push_back(6);
V1.push_back(7);
if ( V2 == V1) // test for equality
cout<< “The vectors V1 and V2 are equal.”<<endl;
else
cout<<“Vectors V1 and V2 are not equal”<<endl;

V2 = V1; // V1 is assigned to V2
cout<<“After assignment,” << end1;
if( V1!= V2)
cout <<“Vector V1 is not equal to V2”<<endl;
❖ 544 ❖ Programming with C++
else

cout<<“Vector V1 and V2 are equal.”<<endl;

cout<< “The elements of V2 are as below”<<endl;

for ( int i = 0; i< 5; i++)

cout <<“ V2[“<<i<<“] = “<< V2[i]<<endl;

return 0;
}

The expected output of the above program is as below.

Vectors V1 and V2 are not equal
After assignment,
Vector V1 and V2 are equal.
The elements of V2 are as below
V2[0] = 2
V2[1] = 4
V2[2] = 5
V2[3] = 6
V2[4] = 7
FUNCTION INSERT()

The function insert() takes three arguments, the first is the location where the insertion
is desired, second is the number of insertions and the third is the value of each insertion. See
the following program for illustration.

PROGRAM 22.8 – Illustrates insertion of additional elements at specified locations in

a vector.

#include<iostream>
#include<vector>
#include<cmath>
using namespace std;
vector<int> V1;
vector<double> V2 ;
vector<char> V3;
void main()
{ for ( int i =0; i<=4;i++)
{V1.push_back (i+2 ); // creating a vector
V2. push_back (sqrt(i+2));
V3.push_back ( 65 + i); }
vector<int>::iterator K1 = V1.begin();
vector<double>::iterator K2 = V2.begin();
vector<char>::iterator K3 = V3.begin();
 Sequence Containers – vector, list and deque ❖ 545❖
cout <<“Before insertion.”<<endl;
cout << “V1\tV2\t V3”<<endl;
while ( K1 != V1.end() )
{cout <<*K1 <<“\t”<< *K2 <<“\t “ << *K3<< endl;
K1++;
K2++;
K3++;
}
vector<int>::iterator A1 = V1.begin();
vector<double>::iterator A2 = V2.begin();
vector<char>::iterator A3 = V3.begin();
V1.insert(A1 ,3,9); //insert three 9 at the beginning of V1.
V2.insert(A2+2,3,10.5); /* insert three elements each of
value 10.5 at 3rd location.*/
V3.insert (A3+3,3,’A’); /* insert 3 elements each equal to A
at the 4th location.*/
cout<< “ After insertion.”<<endl;
cout << “V1\tV2\t V3”<<endl;
while ( A1 != V1.end() )
{cout <<*A1 <<“\t”<< *A2 <<“\t “ << *A3<< endl;
A1++;
A2++;
A3++;
}}

Before the insertion V1 has elements 2, 3, 4, 5 and 6. V2 has elements which are square
roots of elements of V1 and V3 has elements A, B, C, D and E. In V1 three elements each
of value 9 have been inserted at the beginning, in V2 three elements each of value 10.5 have
been inserted at the third location in V2. Similarly three A have been inserted at the fourth
place in V4. The output is as below.
Before insertion.
V1 V2 V3
2 1.41421 A
3 1.73205 B
4 2 C
5 2.23607 D
6 2.44949 E
After insertion.
V1 V2 V3
9 1.41421 A
9 1.73205 B
9 10.5 C
2 10.5 A
❖ 546 ❖ Programming with C++
3 10.5 A
4 2 A
5 2.23607 D
6 2.44949 E

CONTAINER CLASS LIST

22.7 FUNCTIONS AND OPERATORS SUPPORTED BY LIST CLASS

The list is another sequential container class in C++. The list class allows fast insertion and
deletions anywhere in a list. List supports only bidirectional iterators. Many of the member
functions of list class are similar to that of vector class with a few additions. The functions
common to both vectors and lists are given in Table 22.3.
Table 22.3 – Functions common to the vector and list classes
assign() back() begin() clear()
empty() end() erase() front()
insert() max_size() pop_back() push_back()
rbegin() remove() rend() resize()
size() swap()

Table 22.4 – Additional Functions in class list

Function Description
merge() Merges a list with another.
pop_front() The function removes first element of list.
push_front() The function adds an element to the front of list.
remove () The function removes the element from the list.
remove_if() Removes the element if a predicate is satisfied.
reverse () The function reverses the list.
sort() Sorts the list in ascending order.
splice() The function removes elements from one list to another.
unique() The function removes consecutive duplicate elements. The list
should first be sorted to bring make duplicate elements
consecutive.

The codes for application of the functions given in Table 22.3 are similar to those already
illustrated in case of vector class, however, the following programs give the illustrations of
applications of the additional functions described in Table 22.4 along with some of the ones
 Sequence Containers – vector, list and deque ❖ 547❖
described in Table 22.3. The list supports the relational operators. A list can be assigned with
operator = and compared with other lists with overloaded comparison operators >, <, == ,
!=, >=, <= , etc. The following program illustrates the construction of lists by different
methods.

PROGRAM 22.9 – Illustrates the application of functions assign(), push_back(),

begin (), end (), and operators = and == for lists.

#include <iostream>
#include <list>
#include <string>
using namespace std;
list <int> L1 , L2 ,L3 ; // Declares three lists of type int.
list <string> L4; //Declares list of strings.
void main()
{
for (int i=1; i<6;i++)
L1.push_back(i*i ); // putting in values by push_back
L2.assign (5, 75); // Putting values by assignment
L3 = L1; // construction of list by assigning.
const int n =5;
string Name[n]={ “Kunti”, “Sunita”,”Preeti”, “Kamla”, “Roshan”};
for(int j=0; j<5; j++)
L4.push_back( Name[j] );
// pushing values in form of strings
cout<<“The original lists are as below.”<<endl;
cout<< “L1 = ” ;
list<int>::iterator itr ; //Declaration of iterator
for(itr = L1.begin(); itr !=L1.end(); itr ++)
cout<< *itr <<“ ”;
cout<<“\n”;
cout<<“L2 = ”;
for(itr = L2.begin(); itr !=L2.end(); itr ++)
cout<<*itr <<“ ”;
L3 = L2;
cout<<“\n”;
cout<<“L3 = ”;
for(itr = L3.begin(); itr !=L3.end(); itr ++)
cout<<*itr<<“ ”;
cout<<“\nL4 = ”;
❖ 548 ❖ Programming with C++
list<string>::iterator iter ;
for(iter = L4.begin(); iter !=L4.end(); iter ++)
cout<< *iter <<“ ”;
L3 = L1 ; // L1 is copied on to L3
cout<<“\nNew L3 = ”;
for(itr = L3.begin(); itr !=L3.end(); itr ++)
cout<<*itr<<“ ”;
if ( L1 == L3) // comparison of two strings.
cout<<“\nL1 and L3 are now equal”<<endl;
else
cout<<“\nL1 and L3 are not equal\n”;
}

The expected output is given below.

The original lists are as below.
L1 = 1 4 9 16 25
L2 = 75 75 75 75 75
L3 = 75 75 75 75 75
L4 = Kunti Sunita Preeti Kamla Roshan
New L3 = 1 4 9 16 25
L1 and L3 are now equal

22.8 APPLICATION OF SOME FUNCTIONS OF LIST CLASS

Functions unique(), reverse() and remove()

The function unique removes the duplicate elements from the list. The function reverse()
reverses the order of elements in the list, i.e. the first element becomes the last element with
similar changes in other elements. See the following program for illustration.

PROGRAM 22.10 – Illustrates functions unique(), reverse() and remove() for list.

#include<iostream>
#include<list>
using namespace std;
list<char> L1 ;
void main()
{
for (int i=0; i<6;i++)
{ L1.push_back(73 - i );
L1.push_back(73 - i );
}
cout <<“The original list is as below.”<<endl;
list<char>::iterator T1;
 Sequence Containers – vector, list and deque ❖ 549❖
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
cout <<endl;
L1.unique(); // remove the duplicate copies.
cout <<“ After removal of duplicate copies the list is:”<<endl;
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
L1.reverse();
cout<<“\n After reversing the list it is \n”;
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
cout <<“\nAfter removal of F the list becomes \n”;
L1.remove (‘F’);
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
cout <<endl;
}

The expected output is given below.

The original list is as below.
I I H H G G F F E E D D
After removal of duplicate copies the list is:
I H G F E D
After reversing the list it is
D E F G H I
After removal of F the list becomes
D E G H I

The output of the above program is self explanatory. The second line of output is the
original list which has duplicate elements. With function unique the duplicate elements are
removed and the resultant list consists of unique values (4th line of output). The list is then
reversed. In the reversed list the letter F is removed with function remove(). The last line
of output is the resulting list.

PROGRAM 22.11 – Illustrates functions begin(),end()and functions swap() and

sort() for list.

#include<iostream>
#include<list>
using namespace std;
list<char> L1 ;
list<int> L2 ;
❖ 550 ❖ Programming with C++
list <int> L3;
void main()
{
int A [] = {78, 58, 90, 44, 33,11} ;
for (int i=0; i<6;i++)
{L1.push_back(75 - i );
L2.push_back (A [i]);
L3. push_back(i*i);
}
cout<<“The original lists are as below.”<<endl;
list<char>::iterator T1;
list <int>::iterator T2;
cout<< “List L1 = ”;
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1<<“ ” ;
cout<<“\nList L2 = ”;
for ( T2 = L2.begin() ; T2!=L2.end() ; T2++ )
cout <<“ ” <<*T2<<“ ” ;
cout<<“\nList L3 = ”;
for ( T2 = L3.begin() ; T2!=L3.end() ; T2++ )
cout <<“ ” <<*T2<<“ ” ;
L3.sort();
L2.sort();
L2.swap(L3); // swapping elements of two lists
cout<<“\nThe sorted and swapped lists are as below.”<<endl;
cout<< “\nList L2 =” ;
for ( T2 = L2.begin() ; T2!= L2.end() ; T2++)
cout <<“ ” <<*T2<<“ ” ;
cout<< “\nList L3 =” ;
for ( T2 = L3.begin() ; T2!= L3.end() ; T2++)
cout <<“ ” <<*T2<<“ ” ;
}

The expected output is given below.

The original lists are as below.
List L1 = K J I H G F
List L2 = 78 58 90 44 33 11
List L3 = 0 1 4 9 16 25
The sorted and swapped lists are as below.
List L2 = 0 1 4 9 16 25
List L3 = 11 33 44 58 78 90
 Sequence Containers – vector, list and deque ❖ 551❖
In the above program the list L2 and list L3 are swapped and sorted. It is clear from the
output of the program.
The following program declares a list of elements of type char and by name L1 and
another by name L2 and of type int. The list L1 is constructed by function push_back ().
L2 is assigned 5 elements each equal to 50 by the function assign(). The function
L1.front () and L1.back() give reference to the first and the last element of the list
L1 respectively. The two lists are displayed with the help of iterators.

PROGRAM 22.12 – Illustrates functions front(),back(),begin(),end() for lists.

#include<iostream>
#include<list>
using namespace std;
list <char> L1 ;
list<int> L2;
int main()
{
for (int i=0; i<5;i++)
{ L1.push_back(65+i);
L2.assign( 5, 50 ); }
list <char> ::iterator T1= L1.begin() ;
cout<<“The list L1 is.”<<endl;
while ( T1 !=L1.end())
{cout <<“ ”<<*T1 <<“ ”;
T1++; }
cout<<“\n”;
list <int> :: iterator T2 = L2.begin();
cout<< “The list L2 is. ” <<endl;
while
( T2 !=L2.end())
{
cout <<“ ”<< *T2 <<“ ”;
T2++;}
cout<<“\n”;
cout <<“The first element of L1 is “<<L1.front()<<endl;
cout<<“The last element of L1 is “ <<L1.back()<<endl;
return 0;
}
❖ 552 ❖ Programming with C++
The expected output of the program is given below.
The list L1 is.
A B C D E
The list L2 is.
50 50 50 50 50
The first element of L1 is A
The last element of L1 is E

FUNCTION MERGE()

If the list L2 is required to be merged with list L1, the code is written as below.
L1.merge(L2);
After the execution of the program the list L2 becomes empty. This is illustrated in the
following program.

PROGRAM 22.13 – Illustrates application of function merge() for list.

#include<iostream>
#include<list>
using namespace std;
list<int> L1,L2 ;
void main()
{
for (int i=1; i<5;i++)
{ L1.push_back(i);
L2.push_back(i*i);
}
cout<<“The lists are given below.”<<endl;
cout<<“ List1\t List2 ”<<endl;
list<int>::iterator T1;
list<int>::iterator T2;

for(T1=L1.begin(),T2=L2.begin(); T1 != L1.end() ;T1++,T2++ )

cout<<“ ”<<*T1 <<“ \t ”<< *T2 <<endl;
L1.merge(L2); // merge L2 with L1
cout<<“New list1:”<< endl;
for (T1=L1.begin() ; T1 != L1.end(); T1++ )
cout<<“ ” <<*T1 <<“ ” ;
if ( L2.empty())
cout<< “\nL2 is now empty.”<<endl;
}
 Sequence Containers – vector, list and deque ❖ 553❖
The expected output is given below.
The lists are given below.
List1 List2
1 1
2 4
3 9
4 16
New list1:
1 1 2 3 4 4 9 16
L2 is now empty.

DEQUE
The class deque, also called double ended queue, supports random iterator. The components
are stored in adjacent memory blocks. Individual element may be accessed by random iterator
or by [ ] operator like an array. Elements may be added or removed at both ends of a deque.
The salient features of deque as compared with those of list and vector are given in Table
22.1. This container has a combination of some characteristics of both vectors as well as lists.
The functions of class deque are listed in Table 22.5.

22.9 FUNCTIONS AND OPERATORS SUPPORTED BY DEQUE

Some of the commonly used functions supported by this class are listed in Table 22.5. The
functions supported by deque in addition to those for vectors are push_front () and
pop_front() while some functions like capacity() and reserve() are not members
of class deque.
Table 22.5 – Functions supported by deque class
assign () at() back () begin() clear ()
empty () end () erase () front () insert ()
max_size() pop_back() push_back () pop_front push_front
rbegin() rend() resize () size() swap()

Operators Supported by deque class. The deque class supports the relational operators
such as == , != , >=, <= , >, < and = .

PROGRAM 22.14 – Illustrates iterator, push_back(), push_front() for deque.

#include<iostream>
#include<deque>
using namespace std;
deque<int> Q1, Q3;
deque <char> Q2;
❖ 554 ❖ Programming with C++
int main()
{for ( int i =0; i<4;i++)
{Q1.push_back (i*i);
Q2.push_back (65 +i);
}
Q3 = Q1; // Q1 is assigned to Q3
Q3.push_front (20);
Q3.push_back (20*20);
for ( int j = 0; j<4; j++)
cout << Q1[j]<<“\t”<<Q2[j] <<endl;
cout<< “\nSize of Q3 = ” << Q3.size()<<endl;
cout << “Elements of Q3 are : ”;
deque<int>:: iterator iter;
for (iter = Q3.begin() ; iter < Q3.end(); iter++)
cout<< *iter <<“ ”;
return 0;
}

The expected output is given below.

0 A
1 B
4 C
9 D
Size of Q3 = 6
Elements of Q3 are : 20 0 1 4 9 400

22.10 APPLICATION OF SOME FUNCTIONS SUPPORTED BY DEQUE

PROGRAM 22.15 – Illustrates application of push_back(), pop_front()and

pop_back()and size() for deque.

#include<iostream>
# include <deque>
using namespace std;
int main()
{ deque <double> Q1;
for ( int i =0; i< 7;i++)
Q1.push_back (1.5*i);
cout<<“The elements of queue are, ”<<endl;
for ( int j = 0; j<6; j++)
 Sequence Containers – vector, list and deque ❖ 555❖
cout <<“ ”<< Q1[j]<<“ ” ;
cout<<endl;

cout << “The size of Q1 is ” << Q1.size()<<endl;

Q1.pop_back();

Q1.pop_front();
cout<<“Now size of Q1 is ”<<Q1.size()<<endl;

deque <double> :: iterator itr; // itr is the name of iterator

Q1.insert((Q1.begin() +2),2,50.5); ;
// insert 2 elements of value 50.5.
cout<<“Elements of Q1 are,”<<endl;
for ( itr = Q1.begin(); itr < Q1.end() ; itr ++)
cout<<*itr <<“ ” ;

cout<<endl;
return 0;
}

The expected output is as below.

The elements of queue are,
0 1.5 3 4.5 6 7.5
The size of Q1 is 7
Now size of Q1 is 5
Elements of Q1 are,
1.5 3 50.5 50.5 4.5 6 7.5

Since deque is double ended queue, elements may be added at the front-end as well as at
the back-end of the queue. The following program illustrates it.

PROGRAM 22.16 –Illustrates application of functions erase(), push_front(),

push_back() and insert() for deque.

#include<iostream>
# include <deque>
using namespace std;
int main()
{ deque <int> Q1;
Q1.assign(4, 12) ;
cout<<“The elements of queue are, ”<<endl;
❖ 556 ❖ Programming with C++
for ( int j = 0; j<4; j++)
cout <<“ ”<< Q1[j]<<“ ” ;
cout<<endl;
cout << “The size of Q1 is ” << Q1.size()<<endl;
cout<< “Put two tens at front and two 20s at end.”<<endl;
Q1.push_front(10 );
Q1.push_front(10 );
Q1.push_back(20);
Q1.push_back(20);
deque <int> :: iterator itr ;
// itr is the name of iterator
for ( itr= Q1.begin(); itr != Q1.end(); itr++)
cout << *(itr ) <<“ ” ;
cout << “\nElements 2nd to 4th are erased,”<<endl;
Q1.erase ( (Q1.begin() +1),(Q1.begin() +4 ) );
cout<<“Insert two 50s at 3rd location.”<<endl;
Q1.insert((Q1.begin() +2),2,50 );
cout<<“Now size of Q1 is ”<<Q1.size()<<endl;
cout<<“Elements of Q1 are,”<<endl;
for ( itr = Q1.begin() ; itr < Q1.end () ; itr ++)
cout<<*itr <<“ ”;
cout<<endl;
return 0;
}

The expected output is as below.

The elements of queue are,
12 12 12 12
The size of Q1 is 4
Put two tens at front and two 20s at end.
10 10 12 12 12 12 20 20
Elements 2nd to 4th are erased,
Insert two 50s at 3rd location.
Now size of Q1 is 7
Elements of Q1 are,
10 12 50 50 12 20 20
 Sequence Containers – vector, list and deque ❖ 557❖
EXERCISES
1. What are the sequence containers?
2. What are similarities and differences between vectors and lists?
3. A vector has to have 5 elements. Each element has value 35. How do you construct the
vector?
4. A list has to have 5 names as its elements. How will you declare and initialize the list?
5. What are the main differences between a list and a deque?
6. What is the difference between function capacity () and function size()?
7. What are the arguments of the function erase() of vector class?
8. Which functions are supported by list but not supported by vector?
9. What are the iterators ?
10. How do you declare an iterator for a vector with int elements?
11. What do you understand by function begin() and end ()?
12. What is the difference between the functions back () and end ()?
13. What is the difference between the function begin () and function rbegin ()?
14. What is the difference between function begin() and front() when applied to list?
15. Give examples of code to illustrate function remove () and function remove_if ().
16. What do you understand by function max_size () when applied to deque?
17. Make a program to illustrate the function sort() for sorting a list of 10 characters entered
by a user.
18. Make a program for deque in order to illustrate the functions size(), resize() and
insert().
19. Make a program for vectors to illustrate the different constructors.
20. Make a program in which two vectors of equal size are constructed and swapped and
displayed on the monitor.
21. Make a program for list and illustrate the application of the function insert() any where
in the list.
22. Make a list of characters in the decreasing order from Z to A and sort it in the increasing
order A to Z and display the contents.
Answer:

PROGRAM 22.17 – Sorting a list consisting of characters from Z to A.

#include<iostream>
#include<list>
using namespace std;
list <char> L1; // declaration of list
void main()
{
for (int i=0; i<26;i++)
❖ 558 ❖ Programming with C++
L1.push_back(90 - i ); // construction of list
list<char>::iterator T;
cout<<“ The original list is: ”<<endl;
for ( T = L1.begin(); T!=L1.end(); T++)
cout <<“ “ <<*T ; //output statement
cout<<“\n\n”;
L1.sort() ; // sort function
{list<char> ::iterator IT;
cout<<“ The sorted list is: ”<<endl;
for ( IT = L1.begin(); IT!=L1.end(); IT++)
cout<<“ “ <<*IT ; //output statement after sorting
}
cout<<endl;
}

The expected output is given below.

The original list is:
ZYXWVUTSRQPONMLKJIHGFEDCBA
The sorted list is:
ABCDEFGHIJKLMNOPQRSTUVWXYZ
23. Make a program which illustrates the application of function at() and shows that it is
superior to calling the element by index operator [].

❍❍❍
CHAPTER

23.1 INTRODUCTION

The associative containers support fast retrieval of elements of sets through keys. The container
size may be varied dynamically. The elements of an associative set are of type value_type.
Each element is associated with a key of type key_type. The associative containers comprise
(i) sets, (ii) multisets, (iii) maps and (iv) multimaps. The header file for sets and multisets is
<set> and for map and multimaps is <map>.
In case of sets and multisets the element value itself is the key, i.e. key and the associated
value is one and same. While in case of maps and multimaps the keys are different from the
associated values. Thus in sets and multisets an element comprises single value while in maps
and multimaps an element is a pair comprising a key and a value. It is natural to expect that
key associated with an element is immutable while the element value may be changed. Since
in sets the element value and the key are same, therefore, set elements are not mutable.
The difference in sets and multisets is that in a set there are unique keys (or values because
key and associated value is one and the same) while a multiset may have more than one
element (key) with same value, i.e. multiple copies. Similarly in a map only unique keys are
allowed while in multimaps two or more elements may have same key. However, in maps
or multimaps there are no restrictions on the associated values. Two or more keys in a map
or multimap may have same associated values.
Associative containers keep the keys in a specified order. The default order is ascending
order. In case of sets and multisets the stored values are also the keys so the values are stored
in ascending order by default. But this can be altered by specifying the mode of arranging
the keys.

CONSTRUCTOR FOR ASSOCIATIVE CONTAINERS

The associative container classes have a default constructor, a copy constructor and a destructor.
A default constructor does not take any argument. A copy constructor creates a copy of the
given set. Destructor is called for removing all the elements of a container from the memory
and free the memory.
❖ 560 ❖ Programming with C++

23.2 FUNCTIONS SUPPORTED BY ASSOCIATIVE CONTAINERS

Functions supported by sets, multisets, maps and multimaps are described in the following
table.
Table 23.1

S.No. Function Description

1. begin() It returns iterator to first element of the set.
2. clear() The function deletes all the elements of a set.
3. count() The function returns the number of elements that match a key.
4. empty() The function returns true if there is no element in the set, false other-
wise.
5. end() It returns iterator just past the end of set.
6. equal_range() The function returns two iterators, one to the first element that
matches key and second to the element just past the last element
that matches the key.
7. erase() It deletes element at the indexed location or elements between start
and end iterators or erases all the elements. The start element is
included but the end element is not included.
8. find() The function returns an iterator to key if found, otherwise it returns
iterator to end of set.
9. insert() Inserts a value (element) in a set if the value does not already exist
(ii) inserts a number of elements from start to end.
10. key_comp() The function compares keys.
11. lower_bound() The function returns iterator to the first element that has value
greater than or equal to key.
12. max_size() The function returns the maximum number of elements that the set
can hold.
13. rbegin() The function returns a reverse_iterator to the end of the set.
14. rend() Return a reverse_iterator to the beginning of the set.
15. size() Returns the current number of elements in the set.
16. swap() Exchanges the content of one set with that of another.
17. upper_bound() The function returns iterator to the first element in the set with a
key greater than specified key.
18. value_comp() The function compares values.
 Associative Containers ❖ 561❖

OPERATORS SUPPORTED BY ASSOCIATIVE CONTAINERS

All the associative sets support comparison operators == , != , > , < , >= , <= besides the
assignment operator = . Thus associative containers can be assigned as well as compared.

23.3 THE SETS

General characteristics of sets and other associative containers have already been described above.
In the following few programs the operation of some of the functions listed in Table 23.1
and operators supported by sets are illustrated. The iterators are used for output of elements
of sets. The program given below is an illustration of construction and output of elements
of sets.

PROGRAM 23.1 – Illustrates construction of sets and functions begin() and end().

#include<iostream>
using namespace std;
#include <set>
#include <iterator>
const int n = 8,m = 5;
int main()
{ char ch [m] = { ‘A’,’D’,’E’ ,’C’, ‘B’ };
int AR1[n] = { 65, 72, 67, 69, 71, 68, 66, 70 };
set <int> Set1( AR1, AR1+7); // set constructed with an array
set <char> Set3 (ch, ch +5); // set constructed with an array
set<int> Set2 (Set1); // Set2 is constructed as copy of Set1
set<int>:: iterator itr; // declaration of iterator for int set
cout<<“Set1 elements are: ”;
//below is the output statement for set1
for( itr = Set1.begin(); itr!= Set1.end(); itr++)
cout<< *itr<<“ ”;
cout<<“\nSet2 elements are :”;
// below is the output statement for set2
for( itr = Set2.begin(); itr!= Set2.end(); itr++)
cout<< *itr<<“ ”;
cout<<“\nSet3 elements are :” ;
set<char>:: iterator iter; // declaration of iterator
for( iter = Set3.begin(); iter!= Set3.end(); iter++)
cout<< *iter<<“ ”; // output statement for set3
return 0;}
❖ 562 ❖ Programming with C++

The expected output is given below. The output is self explanatory.

Set1 elements are: 65 66 67 68 69 71 72
Set2 elements are :65 66 67 68 69 71 72
Set3 elements are :A B C D E
The output shows that the values have been ordered in an ascending order on entering a
set. The corresponding array elements are not in any order.
The following program illustrates the application of functions size() which returns the
current number of elements in the set, lower_bound() which return iterator to a value just
above the value specified in lower bound, swap() which exchanges the values in the two
sets, clear() which removes all the elements of a set and empty() which checks if there are
any elements in the set.

PROGRAM 23.2 – Illustrates application of size(), lower_bound(), swap()

and empty() functions for sets.

#include<iostream>
using namespace std;
#include <set>
int main()
{ int G[8] = { 100, 78, 34, 34, 34 ,20 ,104,78 };
set <int> S1 ( G, G+8) ; // Elements G to G+8 are entered
set <int> S2 (G, G+5) ; // Elements G to G+5 are entered
cout << “Size of S1 = ”<<S1.size()<<endl ;
cout << “Size of S2 = ”<<S2.size()<<endl;
cout<<“number of element of value 34 = ”<<S1.count(34)<<endl;
cout<<“Lower_bound of 35 in S1 = ”<< *(S1.lower_bound(35))<<endl;
cout <<“The elements of S1 are : ”;
set<int> :: iterator iter;
for (iter = S1.begin(); iter != S1.end(); iter++)
cout<<*iter <<“ ”;
cout<< “\nThe elements of S2 are : ”;
for (iter = S2.begin(); iter != S2.end(); iter++)
cout<<*iter <<“ ”;
S2.swap(S1); // swap S2 with S1
cout <<“\nAfter swapping\n”;
cout <<“ The elements of S1 are : ”;
for ( iter = S1.begin(); iter != S1.end(); iter++)
cout<<*iter <<“ ”;
cout<< “\nThe elements of S2 are : ”;
for ( iter = S2.begin(); iter != S2.end(); iter++)
 Associative Containers ❖ 563❖
cout<<*iter <<“ ”;
S2.clear();
if (S2.empty()) // test if there is any element in the set.
cout<<“\nAfter operation of clear(), S2 is empty.\n”;
return 0;
}

The expected output is given below.

Size of S1 = 5 // Note that duplicate elements are removed
Size of S2 = 3
number of element of value 34 = 1 // duplicates are removed
Lower_bound of 35 in S1 = 78
The elements of S1 are : 20 34 78 100 104
The elements of S2 are : 34 78 100
After swapping
The elements of S1 are : 34 78 100
The elements of S2 are : 20 34 78 100 104
After operation of clear(), S2 is empty.

From the output it is clear the elements in a set are arranged in ascending order. That is
the default setting. Also a set does not contain duplicate elements. The function lower_bound
() returns iterator to an element just above the specified key. The value of lower_bound is
obtained by dereference operator (*). In the following program the descending order is
specified.

PROGRAM 23.3 – Illustrates arranging the set in a specified order.

#include<iostream>
using namespace std;
#include <set>
typedef std::set<int , std::greater <int> > Inset;
// std:: greater <int> for descending order
// std::less for ascending order.
int main()
{int G[8] = { 100, 78, 34, 34, 34 ,20 ,104,78 };
Inset S1( G, G+8) ;
Inset S2 (G, G+5) ;
cout << “Size of S1 = ”<<S1.size()<<endl ;
cout << “Size of S2 = ”<<S2.size()<<endl;
cout<<“number of element of value 34 = ”<<S1.count(34)<<endl;
cout<<“Lower_bound of 35 in S1 = ”<< *(S1.lower_bound(35))<<endl;
cout <<“ The elements of S1 are : ”;
Inset :: iterator iter;
❖ 564 ❖ Programming with C++
for ( iter = S1.begin(); iter != S1.end(); iter++)
cout<<*iter <<“ ”;

cout<< “\nThe elements of S2 are : ”;

for ( iter = S2.begin(); iter != S2.end(); iter++)
cout<<*iter <<“ ”; // for output of elements
S2.clear();
if (S2.empty())
cout<<“\nS2 is empty after operation of clear().”;
return 0; }

The expected output is given below. S1 is now arranged in descending order.

Size of S1 = 5
Size of S2 = 3
number of element of value 34 = 1
Lower_bound of 35 in S1 = 34
The elements of S1 are : 104 100 78 34 20
The elements of S2 are : 100 78 34
S2 is empty after operation of clear().

The following program illustrates creating sets of strings (class objects). An array of strings
consists of names. This array on putting into a set gets arranged in ascending order (A to Z)
by the default setting.

PROGRAM 23.4 – Illustrates sets consisting of objects of class string and characters.

#include<iostream>
using namespace std;
#include <string>
#include <set>
const int n =4;
int main()
{string Str[n] = {“Sunita”,”Asha”, “Priti”, “Kishor” };
char ch[n] = { ‘B’, ‘C’, ‘A’, ‘V’} ; // char array
set <string> S1 ( Str, Str+n);
set <char> S2 (ch, ch+n) ;
cout << “Size of S1 = ”<<S1.size()<<endl ;
cout << “Size of S2 = ”<<S2.size()<<endl;
cout <<“The elements of S1 are : ”;
set<string> :: iterator iter;
for ( iter = S1.begin(); iter != S1.end(); iter++)
cout<<*iter <<“ ”;
cout<<“\nAfter erase elements of S1 are: ”;
 Associative Containers ❖ 565❖
S1.erase( S1.begin()); // erase the first element
for ( iter = S1.begin(); iter != S1.end(); iter++)
cout<<*iter <<“ ”;
set<char> :: iterator itr;
cout<< “\nThe elements of S2 are : ”;
for ( itr = S2.begin(); itr != S2.end(); itr++)
cout<<*itr <<“ ”;
S2.clear();
if(S2.empty())
cout<<“\nS2 is empty”<<endl ;
return 0;
}

The expected output is given below. Note the names are arranged in alphabetical order in
the set.
Size of S1 = 4
Size of S2 = 4
The elements of S1 are : Asha Kishor Priti Sunita
After erase elements of S1 are: Kishor Priti Sunita
The elements of S2 are : A B C V
S2 is empty

In the following program the function erase()is used to remove the first element in
the set and function insert()is used to insert a character in the set.

PROGRAM 23.5 – Illustrates erase() and insert() functions for sets.

#include<iostream>
using namespace std;
#include <set>
#include <iterator>
typedef std::set< char> Chrset; // typedef for avoiding
// writing std::set< char> many times
int main()
{const int n =4;
char G[n] = { ‘Z’, ‘S’ , ‘N’, ‘A’ };
Chrset Set1 (G , G+n); // construction of set
Chrset Set2 (Set1);
Chrset:: iterator itr; // declaration of iterator
cout<<“Set1 elements are : ”;
for( itr = Set1.begin(); itr!= Set1.end(); itr++)
cout<< *itr<<“ ”;
cout<<“\n”;
❖ 566 ❖ Programming with C++
cout<<“Set2 elements are : ”;
for( itr = Set2.begin(); itr != Set2.end(); itr++)
cout<< *itr<<“ ”;
Set1.erase( Set1.begin()); // erases the first element
cout<<“\n”;
Chrset:: iterator iter;
Set1.insert(‘T’); // insertion of an element
cout<< “After insertion Set1 elements are: ”;
for( iter = Set1.begin(); iter!= Set1.end(); iter++)
cout<< *iter<<“ ”;
return 0;
}

The expected output is given below. It is easily compared with the code statements.
Set1 elements are : A N S Z
Set2 elements are : A N S Z
After insertion Set1 elements are: N S T Z

23.4 THE MULTISETS

Multisets allow multiple elements having same key. The keys are sorted according to the
prescribed order, ascending order is the default order. The following program is an illustration
of a multiset which also illustrates the functions count(), size() and upper_bound(). Also
descending order is specified for sorting of the multiset. For multisets also we have to include
the header file <set> in the program.

PROGRAM 23.6 – Illustrates multisets, specification of order of elements and operation

of functions count(), size() and upper_bound().

#include <iostream>
#include<set> // Header file is same for sets and multisets
using namespace std;
typedef std ::multiset<int, greater <int> > IntMset ;
// greater for descending order and less for ascending order.
int main()
{ const int n = 8;
int bill[n] = {20,50,10,15, 70,10, 40,10 };
IntMset Xset( bill , bill+n); // construction of set
IntMset ::iterator iter; // declaration of iterator.
cout<< “Size of Xset = ” << Xset.size()<<endl;
cout<< “Elements in Xset are : ”;
 Associative Containers ❖ 567❖
for ( iter = Xset.begin(); iter != Xset.end(); iter++)
cout<< *iter<<“ ” ;
cout<<“\nUpper_bound = ”<<*(Xset.upper_bound(50))<<endl;
cout<<“Number of copies of 10 in Xset are = ”<< Xset.count (10)<<
endl;
return 0;
}

The expected output is given below.

Size of Xset = 8
Elements in Xset are : 70 50 40 20 15 10 10 10
Upper_bound = 40
Number of copies of 10 in Xset are = 3

23.5 THE MAPS

In maps and multimaps an element comprises a pair of values. The first one is the key and
the second is the associated value. The keys are sorted according to the prescribed order. The
default order is ascending order. The map class has a default constructor, a copy constructor
and a destructor. The header file is <map> for both maps and multimaps. As already mentioned
in the introduction, keys are not mutable, however, the values associated with the keys may
be changed. The following program is an illustration of map in which the keys are the names
of three metals, i.e. steel, copper and brass, and the associative values are the prices per kg.
After entering the data the prices are revised to new values, i.e. the associated values are changed.

PROGRAM 23.7 – Illustrates construction of a map.

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef map <string, int > Mint ;
// typedef has been used to avoid the
// the writing of map <string, int> several times
int main()
{
Mint Pricelist ;
string Metal[3] = {“steel”, “copper”, “brass”};
int Values [3] = { 20, 200, 180};
for ( int j = 0; j<3; j++)
Pricelist [Metal[j]] = Values[j] ; // input to map
Mint :: iterator iter; // declaration of iterator.
❖ 568 ❖ Programming with C++
cout << “The prices are as below”<<endl;
for ( iter= Pricelist.begin();iter != Pricelist.end(); iter++)
cout<< (*iter).first<<“\t”<<(*iter).second<<endl;
iter = Pricelist.begin();
(*iter).second = 250; // Changing the associated value
(*(++iter)).second = 350; // Changing the associated value
(*(++iter)).second = 25 ; // Changing the associated value
cout<< “Prices after revision are as below”<<endl;
for ( iter= Pricelist.begin();iter != Pricelist.end(); iter++)
cout<< (*iter).first<<“\t”<<(*iter).second<<endl;
return 0;
}

The prices are as below

brass 180
copper 200
steel 20
Prices after revision are as below
brass 250
copper 350
steel 25

In the following program, the input from keyboard has been used to construct the map.
It makes a grade list.

PROGRAM 23.8 – Illustrates input from keyboard to construct a map.

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef map <string, int > Map_Grade ;
int main()
{
string Name ;
int Marks;
Map_Grade Grade; // Grade is name of map
for (int i = 0; i<4; i++) // loop for entering names and marks
{ cout<<“Enter a name : ”;cin>> Name ;
cout<<“Enter the marks obtained : ”; cin >> Marks;
Grade [Name] = Marks ;}
 Associative Containers ❖ 569❖
// below another name and marks are inserted.
Grade.insert (Map_Grade::value_type(“Sujata”, 85));
// inserting an entry of name and value
cout<< “ The grade list is as below.\n\n” ;
Map_Grade :: iterator iter; // declaration of iterator
iter = Grade.begin();
// below is loop for output
for( iter = Grade.begin(); iter != Grade.end(); iter++)
cout<< (*iter).first<<“\t” <<(*iter).second<<“\n” ;
return 0;
}

The expected output is given below.

Enter a name : Manjri
Enter the marks obtained : 80
Enter a name : Sunita
Enter the marks obtained : 79
Enter a name : Bandal
Enter the marks obtained : 85
Enter a name : Chanan
Enter the marks obtained : 80
The grade list is as below.

Bandal 85
Chanan 80
Manjri 80
Sujata 85
Sunita 79

Note that in the output the names (keys) are arranged in alphabetical order. In the
following program the application of functions insert() and function size() are
illustrated.

PROGRAM 23.9 – Application of functions insert() and size().

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef map <string, int > Map_Grade ;
int main()
{ string Name ;
❖ 570 ❖ Programming with C++
int Marks;
Map_Grade Grade;
for (int i = 0; i<4; i++)
{cin>> Name ;
cin >> Marks;
Grade [Name] = Marks ;}
Grade.insert (Map_Grade::value_type (“Sujata”, 85));
Grade.insert (Map_Grade ::value_type (“Mumta” ,90));
cout<<“\n”<< Grade.size ( )<<endl; // number of elements
Map_Grade :: iterator iter;
for( iter = Grade.begin() ; iter != Grade.end(); iter++)
cout<< (*iter).first<<“ \t ” <<(*iter).second<<“\n” ;
return 0;
}

The expected output is as below.

Nany
90
Mini
88
Simi
77
Tiny
85
6
Mini 88
Mumta 90
Nany 90
Simi 77
Sujata 85
Tiny 85

In the following program map is created as a copy of another map.

PROGRAM 23.10 – Use of assignment operator (=) in order to get a copy of map.

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef map <string, int > Map_Grade ;
 Associative Containers ❖ 571❖
int main()
{
string Name ;
int Marks;
Map_Grade Grade1 ;
for (int i = 0; i<4; i++)
{cin>> Name ;
cin >> Marks;
Grade1 [Name] = Marks ;}
Map_Grade Grade2;
Grade2 = Grade1 ; // Application of assignment operator
// to maps
Grade1.insert (Map_Grade::value_type(“Sujata”, 85));
Map_Grade :: iterator iter;
iter = Grade1.begin();
cout<<“The Grade1 list is as below.”<<endl;
for( iter = Grade1.begin(); iter != Grade1.end(); iter++)
cout<< (*iter).first<<“ \t ” <<(*iter).second<<“\n” ;
Map_Grade :: iterator iter2;
cout << “The Grade2 list is as follows.”<<endl;
iter2 = Grade2.begin();
for( iter2 = Grade2.begin(); iter2 != Grade2.end(); iter2++)
cout<< (*iter2).first<<“ \t ” <<(*iter2).second<<“\n” ;
return 0;
}

The expected output is as under.

Kumar
88
Nimi
90
Sunita
78
Trini
77
The Grade1 list is as below.
Kumar 88
Nimi 90
Sujata 85
❖ 572 ❖ Programming with C++
Sunita 78
Trini 77
The Grade2 list is as follows.
Kumar 88
Nimi 90
Sunita 78
Trini 77
The element (“Sujata”,85) was inserted in Grade1 after the assignment. Therefore, the
output of Grade2 does not contain this name.
In the following program the function insert() is used to create a map.

PROGRAM 23.11 – It is another example of application of function insert(). In fact the

entire map is constructed by insert function.

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef std:: map < string ,int > Multigrade ;
int main()
{
Multigrade pairs;
pairs.insert (Multigrade::value_type(“Sunita”, 80));
pairs.insert (Multigrade::value_type(“Sunita”, 85));
pairs.insert (Multigrade::value_type(“Kunti”, 96));
pairs.insert (Multigrade::value_type(“Punita”, 70));
pairs.insert (Multigrade::value_type(“Punita”, 65));
pairs.insert (Multigrade::value_type(“Bubita”, 95));
pairs.insert (Multigrade::value_type(“Mumta”, 60));
pairs.insert (Multigrade::value_type(“Mumtaz”, 60)) ;
cout<< “Present size of pairs is = ” <<pairs.size()<<endl;
Multigrade :: const_iterator itr= pairs.begin();
cout<<“ The contents of pairs are as below.”<<endl;
for(itr = pairs.begin(); itr != pairs.end(); itr++)
cout<< itr-> first <<“\t”<< itr-> second<<“\n”;
pairs.erase (pairs.begin() ); // erase the first element
cout <<“The size after erasing first pair is = “<< pairs.size() <<endl;
cout << “The contents after erase are:”<<endl;
 Associative Containers ❖ 573❖
Multigrade :: iterator iter; // declaration of iterator
for(iter= pairs.begin(); iter != pairs.end(); iter++)
cout<< iter-> first <<“\t”<< iter-> second<<“\n”;
return 0;
}

Output is given below. The duplicate names have been ignored in construction of the map.
Present size of pairs is = 6
The contents of pairs are as below.
Bubita 95
Kunti 96
Mumta 60
Mumtaz 60
Punita 70
Sunita 80
The size after erasing first pair is = 5
The contents after erase are:
Kunti 96
Mumta 60
Mumtaz 60
Punita 70
Sunita 80

The following program illustrates the application of function erase(), function size
() and function max_size ().

PROGRAM 23.12 –Illustrates the application of size(), max_size() and erase()

in maps.

#include<iostream>
using namespace std;
#include<map>
#include<string>
typedef std:: map < string ,int > Multigrade ;
int main()
{ Multigrade pairs;
pairs.insert (Multigrade::value_type(“Sunita”, 80));
pairs.insert (Multigrade::value_type(“Sunita”, 85));
pairs.insert (Multigrade::value_type(“Kunti”, 96));
pairs.insert (Multigrade::value_type(“Punita”, 70));
pairs.insert (Multigrade::value_type(“Punita”, 65));
❖ 574 ❖ Programming with C++
pairs.insert (Multigrade::value_type(“Bubita”, 95));
pairs.insert (Multigrade::value_type(“Mumta”, 60));
pairs.insert (Multigrade::value_type(“Mumtaz”, 60)) ;
cout<< “Present size of pairs is = ” <<pairs.size()<<endl;
cout<<“Maximum numbers that map can hold = ”<<pairs.max_size() <<endl;
Multigrade :: const_iterator itr= pairs.begin();
cout<<“The pairs in the map are as below.”<<endl;
for( itr= pairs.begin(); itr != pairs.end(); itr++)
cout<< itr-> first <<“\t”<< itr-> second<<“\n”;
pairs.erase (++pairs.begin(), --pairs.end() );
// erase from 2nd to last but one element.
cout <<“The size after erasing first pair is = ”<< pairs.size() <<endl;
cout << “The pairs now in map are given below.”<<endl;
Multigrade :: iterator iter;
for( iter= pairs.begin(); iter != pairs.end(); iter++)
cout<< iter-> first <<“\t”<< iter-> second<<“\n”;
return 0;
}

The expected output is as under.

Present size of pairs is = 6
Maximum numbers that map can hold = 1073741823
The pairs in the map are as below.
Bubita 95
Kunti 96
Mumta 60
Mumtaz 60
Punita 70
Sunita 80
The size after erasing first pair is = 2
The pairs now in map are given below.
Bubita 95
Sunita 80

23.6 THE MULTIMAPS

As already discussed in introduction a multimap allows multiple copies of keys. The header
file for multimaps is also <map>. The following program illustrates it.
 Associative Containers ❖ 575❖
PROGRAM 23.13 – Illustrates construction of a multimap.

#include<iostream>
using namespace std;
#include<map> // Header file is same as for maps.
#include<string>
typedef std::multimap < string ,int > Multigrade ;
// With this typedef we need write only Multigrade instead of
// std::multimap < string ,int > many times
int main()
{
Multigrade pairs;
pairs.insert (Multigrade::value_type(“Sunita”, 80));
pairs.insert (Multigrade::value_type(“Sunita”, 85));
pairs.insert (Multigrade::value_type(“Kunti”, 96));
pairs.insert (Multigrade::value_type(“Punita”, 70));
pairs.insert (Multigrade::value_type(“Punita”, 65));
pairs.insert (Multigrade::value_type(“Anita”, 95));
pairs.insert (Multigrade::value_type(“Mumta”, 60));
for( Multigrade :: const_iterator itr = pairs.begin();
itr != pairs.end(); itr++)
cout<< itr-> first <<“\t”<< itr-> second<<“\n”;
return 0;
}

The expected output is given below.

Anita 95
Kunti 96
Mumta 60
Punita 70
Punita 65
Sunita 80
Sunita 85

The following program illustrates the application of functions count()to find how many
times a particular key appears in the map.

PROGRAM 23.14 – Illustrates the function count() to a multimap.

#include<iostream>
using namespace std;
#include<map>
❖ 576 ❖ Programming with C++
#include<string>
typedef std::multimap <string ,int> Multigrade ;
int main()
{
Multigrade pair ;
pair .insert (Multigrade::value_type(“Sunita”, 80));
pair .insert (Multigrade::value_type(“Sunita”, 85));
pair .insert (Multigrade::value_type(“Kunti”, 96));
pair .insert (Multigrade::value_type(“Punita”, 70));
pair .insert (Multigrade::value_type(“Punita”, 65));
pair .insert (Multigrade::value_type(“Anita”, 95));
pair .insert (Multigrade::value_type(“Mumta”, 60));
cout<<“The number of times Sunita appears = ”<< pair.count (“Sunita”) <<endl;
cout<<“The number of times Mumta appears = ”<< pair.count (“Mumta”)<<endl;
cout<< “The number of times Kanta appears = ”<< pair.count ( “Kanta”)<<endl;
cout<<“The names of students and grades are as below.”<<endl;
for( Multigrade :: const_iterator itr= pair .begin();
itr != pair.end(); itr++)
cout<< itr-> first <<“\t”<< itr-> second<<“\n”;
return 0;
}

The expected output is given below.

The number of times Sunita appears = 2
The number of times Mumta appears = 1
The number of times Kanta appears = 0
The names of students and grades are as below.
Anita 95
Kunti 96
Mumta 60
Punita 70
Punita 65
Sunita 80
Sunita 85
 Associative Containers ❖ 577❖

EXERCISES
1. What is difference between a set and a map?
2. What are the methods of constructing a set?
3. Are the values in a set ordered?
4. How would you write the code for the keys in a set to be ordered in descending order?
5. What is the difference between a set and a multiset?
6. How is a multimap different from a map?
7. If the keys in a multimap are the names of students, how would you make a code for
making list of students with grades, listing names in alphabetical order?
8. In the program of Q.7 how would you add code to determine the first ranker?
9. What are the return values of functions lower_bound() and upper_bound()?
10. What does the function equal_range() do?
11. Make a program for arranging names of students in order of merit.
12. Make a program to construct a multimap from two arrays. Insert two more pairs in the
set.
13. What operators may be used with sets? Make program to illustrate the action of operators
= and >= on the sets.
14. Make a program to illustrate the working of functions key_comp() and upper_bound()
on elements of maps.
15. Make a program to illustrate the functions rbegin() and rend().
16. What does function find() do? Make program to illustrate the action of find() and
function count () in a multimap.
17. What is the return value of max_size in a multimap? Make a program to illustrate the
application of function max_size() and value_comp().
18. Make a program for a multimap in which keys are roll numbers of 10 students from
ME200401 to ME200410 and associated values are the marks obtained by students entered
by user in the range from 0 to 100. The program determines the number of students who
have marks between 0 to 40, between 41 to 60, from 61 to 80 and more than 80.

❍❍❍
❖ 578 ❖ Programming with C++

CHAPTER

24.1 INTRODUCTION

Bit sets are useful in many applications, particularly in real time control systems. In fact, the
various processes in computer system itself are controlled by bit sets. Bit sets are particularly
useful when a choice is to be made out of different hardware devices such as printer, keyboard,
etc., connected to a computer or any other device. With a single bit one can choose one out of
two devices. One device for bit = 0 and second for bit = 1. However, if 0 is taken as default
value and neglected, we can connect to a device with one bit as illustrated in Table 24.1 below.
Table 24.1

Bit value (signal) Device connection

0 default case, no device connected
1 connect device1
Similarly with two bits one can choose one out of four cases. However, if the condition when the
two bits are zero is neglected as a default case, choice may be made for one out of three as given in
Table 24.2 below.
Table 24.2

Bit1 value Bit2 value Device connection

0 0 default case, no device connected
1 0 connect device1
0 1 connect device2
1 1 connect device3

Similarly with 3 bits one can select one out of 8 devices and with 4 bits one out of 16
and so on. In bit sets the data structure comprises sets of bits which are sequences of 0s and
1s. This concept can be used in any other control system as well.
The class bitset has a number of functions for manipulation of bit sets. We have to
include the header file <bitset> in the program for using functions of class bitset. The bitwise
operators are listed in Table 24.3. The bitset functions are briefly described in Table 24.5.
 Bit Sets ❖ 579❖

24.2 CONSTRUCTION OF BIT SETS

Bit sets may be constructed by converting a signed or unsigned integer variable (short int,
long or character) into binary representation as given below.
#include <bitset> // Header file for bit sets
bitset(); //an empty bit set
unsigned int n = 384;
bitset <16> Bset1 ( n ); // bit set desired in 16 bits

Name size Name Number

of class of bit set of bit set to be converted into bits
Fig. 24.1: Construction of a bit set

In the following program we make use of the above code to convert integer numbers
into bit sets.

PROGRAM 24.1 – Illustrates construction of bit sets.

#include<iostream>
#include<bitset>
using namespace std;
void main ()
{short int p = 75;
short k = –75;
unsigned int n = 285;
unsigned long m = 24567854;
bitset<8> Bitset1(p), Bitset2(k), Bitset3(‘B’);
bitset<16> Bitset4(n);
bitset<24> Bitset5 (m);
cout << “Bit set for “<<p <<“ is : “ <<Bitset1<<endl;
cout << “Bit set for “<<k <<“ is : “ <<Bitset2<<endl;
cout << “Bit set for “<<‘B’ <<“ is : “ <<Bitset3<<endl;
cout << “Bit set for “<<n <<“ is : “ <<Bitset4 <<endl;
cout << “Bit set for “<<m <<“ is : “ <<Bitset5 <<endl;
}

The expected output is given below.

Bit set for 75 is : 01001011
Bit set for –75 is : 10110101
Bit set for B is : 01000010
Bit set for 285 is : 0000000100011101
❖ 580 ❖ Programming with C++
Bit set for 24567854 is : 011101101110000000101110
The output of the above program shows how a negative integer is stored in computers.
The bit on extreme left is for sign (0 for +ve and 1 for –ve). The remaining part is the
compliment of bit set for 75 plus 1).
In the above case the bit sets are treated as strings of 0s and 1s. If the output is required
as element by element the following code may be used. In a for loop we generally write the
expression as (int i = 0; i<n ;i++), however, for getting bits we have to reverse
the limits, i.e. for ( int i < Bitset_size –1 ; i >= 0; i–– ). The reason is explained in
the program given below.

PROGRAM 24.2 – Illustrates output of bit set as elements of an array.

#include<iostream>
#include<bitset>
using namespace std;
void main ()
{ unsigned long n = 85;
unsigned long m = 456;
bitset<16> Bitset1(n);
bitset<16> Bitset2(m);
cout << “Bit set for “<<n <<“ is: \n “;
for (int i = (long) Bitset1.size()-1 ; i>= 0 ; i––)
cout << Bitset1[i]<<“ “;
cout << “\nBit set for “<< m<<“ is: \n “;
for ( int j = (long) Bitset2.size()-1; j>=0 ; j––)
cout << Bitset2[j] << “ “ ;
cout <<“\n”;
}

The output is given below.

Bit set for 85 is:
0000000001010101
Bit set for 456 is:
0000000111001000

If the above for loop had been written as

for(i=0; i<(long) Bitset2.size() ; i++)
the output for 85 would have been,
1010101000000000
which is obviously wrong because of trailing zeroes which should have been on left hand side.
 Bit Sets ❖ 581❖

24.3 BITSET CLASS OPERATORS

The bitwise operators are also described in Chapter 4. Here these are included in Table 24.3
for ready reference. These are supported by bitset class.
Table 24.3 – Operators supported by bitset class

Operator symbol Description

| Bitwise OR.
& Bitwise AND.
^ Bitwise XOR.
<< Left shift.
>> Right shift.
|= Bitwise OR and assign.
&= Bitwise AND and assign.
^= Bitwise XOR and assign.
<<= Left shift and assign.
>>= Right shift and assign.
!= Not equal to.
== Equal to.
~ Complement operator, equivalent to function flip().
[] Returns reference to indexed bit.

TRUTH TABLES FOR THE OPERATORS OR, XOR, AND AND COMPLEMENT
In all the operators described above the operations are carried out bit by bit. The truth tables
for some of the operators such as complement (~),OR(|), AND (&) and XOR(^) are
illustrated in Table 24.4 below.
Table 24.4 – Truth table for OR, AND, XOR and complement (~) operators
Bit1 Bit2 Complement BitwiseOR Bitwise AND Bitwise XOR
~ Bit1 Bit1 | Bit2 Bit1 & Bit2 Bit1 ^ Bit2
0 0 1 0 0 0
1 0 0 1 0 1
0 1 1 1 0 1
1 1 0 1 1 0

The following program illustrates the application of some of the operators described above.
❖ 582 ❖ Programming with C++

PROGRAM 24.3 – Illustrates application of bitwise operators.

#include<iostream>
#include<bitset>
using namespace std;
int main ()
{unsigned int n = 185;
unsigned int m = 200;
bitset<12> Bset1(( int ) n);
bitset<12> Bset2(( int) m);
cout << “Bset1 for ”<<n <<“ is = ”<<Bset1<<endl;
cout << “Bset2 for ”<< m<<“ is = ” <<Bset2<<endl;
cout <<“ Bset1 OR Bset2 = ” <<(Bset1 | Bset2)<<endl;
cout << “ Bset1 AND Bset2 = ”<<(Bset1 & Bset2 )<<endl;
cout << “ Bset1 XOR Bset2 = ”<<(Bset1 ^ Bset2 )<<endl;
cout << “ Complement of Bset1 = ”<< ~ Bset1 <<endl;
return 0;
}

The expected output is given below.

Bset1 for 185 is = 000010111001
Bset2 for 200 is = 000011001000
Bset1 OR Bset2 = 000011111001
Bset1 AND Bset2 = 000010001000
Bset1 XOR Bset2 = 000001110001
Complement of Bset1 = 111101000110

In the above program bit sets are generated for two numbers 185 and 200. The third
line of the output gives result of OR operation between the two sets. The operation is carried
out bit by bit according to the truth Table 24.4. Similarly the 4th line of output gives the
result of AND operation. The result of XOR operation is given in the fifth line of the output.
The following program illustrates the application of operators |= , &= , ^= and ~ .

PROGRAM 24.4 – Illustrates application of combination assignment operators for bit

sets.

#include<iostream>
#include<bitset>
using namespace std;
int main ()
{unsigned int n = 185;
 Bit Sets ❖ 583❖
int m = 200;
bitset<12> Bset1(( int ) n);
bitset<12> Bset2(( int) m);
cout << “Original Bset1 for “<<n <<“ is = “<<Bset1<<endl;
cout << “Original Bset2 for “<< m<<“ is = “ <<Bset2<<endl;
cout <<“New Bset1= Bset1 |= Bset2 = “ <<(Bset1 |= Bset2)<<endl;
cout <<“New Bset1= Bset1 &= Bset2 = “<<(Bset1 &= Bset2 )<<endl;
cout <<“New Bset1= Bset1 ^= Bset2 = “<<(Bset1 ^= Bset2 )<<endl;
cout << “Reversed Bset2 = ~ Bset2 =”<< ~ Bset2 <<endl;
return 0;
}

The expected output is given below.

Original Bset1 for 185 is = 000010111001
Original Bset2 for 200 is = 000011001000
New Bset1= Bset1 |= Bset2 = 000011111001
New Bset1= Bset1 &= Bset2 = 000011001000
New Bset1= Bset1 ^= Bset2 = 000000000000
Reversed Bset2 = ~ Bset2 =111100110111

The output given above is self explanatory. The reader should compare the program output
with the manual calculations according to the truth table given in Table 24.4.
The operators left shift and right shift are illustrated in the following program.

PROGRAM 24.5 – Illustrates application of shift operators on bit sets.

#include<iostream>
#include<bitset>
using namespace std;
void main ()
{unsigned int n = 185 ,m = 221 ;
bitset<16> Bset1(n);
bitset<16> Bset2(m);
cout << “Bit set for ”<<n <<“ is = “<< Bset1 <<“ \n” ;
cout <<“After right shift 4 digits”<<endl;
cout<<“ Bset1 = ”<<(Bset1>>=4)<<endl; // shift right and assign
cout <<“Bit set for “<< m<<“ is = “<< Bset2<<endl ;
cout << “After left shift by 3 digits”<<endl;
cout<<“Bset2 = ”<< (Bset2<<=3)<<endl; // shift left and assign
if (Bset1==Bset2) //check for equality
cout<<“ The Bset1 and Bset2 are equal”<<endl;
❖ 584 ❖ Programming with C++
else
cout<<“The Bset1 and Bset2 are not equal”<<endl;
}

The expected output is given below.

Bit set for 185 is = 0000000010111001
After right shift 4 digits
Bset1 = 0000000000001011
Bit set for 221 is = 0000000011011101
After left shift by 3 digits
Bset2 = 0000011011101000
The Bset1 and Bset2 are not equal
Shifting the bit set to right by one place has the effect of integer division by 2. Therefore,
the 4 shifts to right of bit set for 185 is equivalent to integer division by 24 = 16. Therefore,
the division gives 185/ 16 = 11 (integer division). In binary, the result is 0000000000001011.
The bit set for 221 is shifted left by three places and zeroes are substituted in the places
vacated by shift. This is equivalent to multiplying 221 by 23 = 8. Therefore, the resultant bit
set Bset2 represents a number 221 × 8 = 1768 which in binary is equal to 0000011011101000.

24.4 BIT SET CLASS FUNCTIONS

The class bitset in header file <bitset> supports the following functions.
Table 24.5 – Functions supported by class bitset.

Function Description
any () The function return true if any of bits is set, false otherwise.
count () The function returns the number of set bits in a bit set.
flip () The function gives inverse bit by bit of the bit set.
none () The function returns true if none of bits is set.
reset () The function is used to reset a bit to 0.
set () The function is used to set a bit to 1.
size () The function returns the number of bits that the bit set can hold.
test () The function returns the value of a bit.
to_string () The function converts the bit set into a string of bits.
to_ulong () The function returns the integral value of bit set.

PROGRAM 24.6 – Illustrates application of functions count(), any (), none()and

reset().

#include<iostream>
#include<bitset>
 Bit Sets ❖ 585❖
using namespace std;
void main ()
{unsigned int n = 125 ,m = 11765,p=0;

bitset<16> Bset1(n);
bitset<16> Bset2(m);
bitset<16> Bset3(p);

cout <<“Bset3 = “<<Bset3<<endl;

cout<<“Number of set bits in Bset3 = ” <<Bset3.count()<<endl;

if (Bset3.any())

cout<< “yes, some bits are set in Bset3”<<endl;

else
cout<<“ No bit is set Bset3”<<endl;

if (Bset3.none())
cout<< “None of bits are set”<<endl;
else
cout<<“ Some bits are set.” <<endl;

cout << “Bit set for ”<<n <<“ is = ”<< Bset1<<endl ;

Bset1.reset(5);
cout << “After resetting 6th bit it is = ”<< Bset1 <<“ \n”;
cout << “Bit set for ”<< m<<“ is = ”<<Bset2 <<endl ;

cout<<“Number of set bits in Bset2 = ” <<Bset2.count()<<endl;

Bset2.set (3);
cout <<“After setting 4th bit it is = ”<< Bset2 <<endl ;
}

The output of the program is self explanatory and is given below.

Bset3 = 0000000000000000
Number of set bits in Bset3 = 0
No bit is set Bset3
None of bits are set
Bit set for 125 is = 0000000001111101
After resetting 6th bit it is = 0000000001011101
Bit set for 11765 is = 0010110111110101
Number of set bits in Bset2 = 10
After setting 4th bit it is = 0010110111111101

FUNCTIONS SIZE(), TO_ULONG() AND ANY()

The function size() returns the number of bits that the bit set can hold. The function
to_ulong () returns the integral value of bit set. The function any() returns true if any
of bit in the bit set is set otherwise it returns false.
❖ 586 ❖ Programming with C++

PROGRAM 24.7 – Illustrates application of functions size(), to_ulong() and any().

#include<iostream>
#include<bitset>
using namespace std;
void main ()
{ unsigned int n = 1123 ,m = 12758;
bitset<16> Bset1(n);
bitset<16> Bset2(m);
cout << “Bit set for “<<n <<“ is \n”;
cout << Bset1 <<“ \n”;
cout <<“ size of bit set = “<< Bset1.size()<<endl;
cout << Bset2 <<endl ;
cout<<“Number equivalent to Bset2 = “ << Bset2.to_ulong()<< endl;
cout<< Bset2.any()<<endl;
}

The expected output is given below.

Bit set for 1123 is
0000010001100011
size of bit set = 16
0011000111010110
Number equivalent to Bset2 = 12758
1

FUNCTION FLIP () AND NONE()

The function flip gives inverse of bit set bit by bit. A set bit is reset and an unset bit is set.
The function none() returns true if none of the bits in a bit set is set otherwise it returns
false.

PROGRAM 24.8 – Illustrates application of functions flip() and none().

#include<iostream>
#include<bitset>
using namespace std;
void main ()
{
unsigned int n = 1123 ,m = 12758;
bitset<16> Bset1(n);
bitset<16> Bset2(m);
cout << “Bit set for ”<<n <<“ is \n”;
 Bit Sets ❖ 587❖
cout << Bset1 <<“ \n”;
cout <<“ Inverse of bit set = “<< Bset1.flip()<<endl;
cout<< Bset1.none()<<endl;
cout << Bset2 <<endl ;

cout<<“ Are the two bit sets equal?”<<endl;

if ( Bset1 == Bset2)
cout << “Yes”<<endl;
else

cout << “No”<<endl;

cout<<“Are the two bit sets unequal?”<<endl;

if ( Bset1 != Bset2)
cout << “ Yes”<<endl;
else
cout << “No”<<endl;
}

The expected output given below is self explanatory.

Bit set for 1123 is
0000010001100011
Inverse of bit set = 1111101110011100
0
0011000111010110
Are the two bit sets equal?
No
Are the two bit sets unequal?
Yes

FUNCTIONS TEST(), SET() AND RESET ()

The function test() returns the state of bit. The function set() sets the specified bit to
1. The function reset() sets off the bit to zero. See the following program for illustration.

PROGRAM 24.9 – Illustrates application of functions test(), reset() and set().

#include<iostream>
#include<bitset>
using namespace std;
void main ()
{unsigned int n = 125 ,m = 11765;
bitset<16> Bset1(n);
bitset<16> Bset2(m);
❖ 588 ❖ Programming with C++
cout << “Bit set for ”<<n <<“ is Bset1 = ”<< Bset1<<endl;
cout<< “Test bit 5. The 5th bit = ” <<Bset1.test(4)<<“\n”;

Bset1.reset(4);
cout<<“After reset bit 5 = “<< Bset1.test(4)<<endl;
cout << “Bit set for ”<<11765 <<“ is Bset2 = ”<<Bset2 <<endl ;
cout << “Test bit 10. The 10th bit = ” << Bset2.test(9)<<endl;

Bset2.set(9) ;
cout<< “After setting the 10th bit = ” <<Bset2.test(9)<<endl;
}

The expected output is given below.

Bit set for 125 is Bset1 = 0000000001111101
Test bit 5. The 5th bit = 1
After reset bit 5 = 0
Bit set for 11765 is Bset2 = 0010110111110101
Test bit 10. The 10th bit = 0
After setting the 10th bit = 1

EXERCISES
1. What are the bit sets?
2. How are the bit sets useful?
3. How bit sets may be used to activate one out of 6 peripherial devices connected to a
computer?
4. In a bit set which is 8 bit wide only, the 4th bit is desired to be set. Write code to set it?
5. Compare the bit sets for 246 and –246. Explain the method of storing –ve numbers in
computer memory.
6. Make a program to illustrate the action of operators &, ^ and | for bit sets.
7. Make a program to illustrate the action of operators << and >> on bit sets. How do the
above shifts change the values represented by the bit sets?
8. Make a program to illustrate the working of operators |= , &= and ^= on bit sets.
9. Make a program to illustrate the following function of bit set class.
(i) any() (ii) count () (iii) flip()
10. Make a program to illustrate the action of following functions of bit set class.
(i) set () (ii) reset () (iii) test ()
11. Make a program to illustrate the action of following functions. Explain their meaning.
(i) none () (ii) size ()
12. What do you understand by function to_ulong() and to_string() as applied to bit
sets? Make a program to illustrate the action of these functions.

❍❍❍
CHAPTER

25.1 INTRODUCTION

In this chapter we deal with templates functions and algorithms which may be reused by a
programmer in application programs. These were briefly mentioned in the Chapter 21 on
STL. Here we shall deal in some detail. The main advantage of using these algorithms is
that the programmer does not have to invent the wheel again. It saves time and effort.
Moreover, being based on template functions and classes, these can be used for any type of
data, i.e. fundamental type as well as user defined type. Besides, these programs have been
made by experts in C++ and thus assure a degree of robustness as well as efficiency.
The algorithms are independent of container classes, so these are applicable to objects of
container classes as well as to arrays, strings and other data types. All the algorithms except
the numeric ones, are contained in the header file <algorithm>. Therefore, we have to include
header file <algorithm> in the programs in which we are using them. The numeric algorithms
which deal with numerical manipulations are contained in the header file <numeric>. Hence
for using these algorithms the header file <numeric> has to be included. In all there are about
75 algorithms. These are broadly classified into following categories. The actions of the
different algorithms in each category are briefly described in Tables 25.2 to 25.9 and many
of them are illustrated in the following pages.
Table 25.1

Algorithm category Header file

Numeric algorithms <numeric>
Comparison (relational) algorithms <algorithm>
Non-mutating algorithms <algorithm>
Mutating algorithms <algorithm>
Searching and sorting algorithms <algorithm>
Algorithms on sets <algorithm>
Permutation algorithms <algorithm>
Algorithms on heap <algorithm>
❖ 590 ❖ Programming with C++

25.2 CATEGORIES OF ALGORITHMS AND BRIEF DESCRIPTIONS

The different categories of algorithms are listed and briefly described in following tables.

RELATIONAL ALGORITHMS
Table 25.2 – Relational algorithms (Header file <algorithm>)

Algorithm Description
equal() Returns true if two sequences are equal else false.
lexicographical_compare() Compares two sequences lexicographically. It returns
true if the first is less than the second.
max() Returns greater of the two values.
min() Returns smaller of the two values.
max_element() Returns iterator to the largest element in the range.
min_element() Returns iterator to element with minimum value.
mismatch() Finds the first position of mismatch between two
sequences of same size.

NON-MUTATING ALGORITHMS
Table 25.3 – Non-mutating sequence algorithm (Header file <algorithm>)

Algorithm Description
adjacent_find() Finds the first adjacent pair of elements which have same value.
count() Counts the number of elements that match the specified value.
count_if() Counts the number of elements that satisfy a predicate.
find() Finds the first element that matches the specified value.
find_end() Finds the last occurrence of a subsequence.
find_first_of() Finds the first occurrence of an element of a given sequence.
find_if() Finds the first element that satisfies the specified predicate.
for_each() Applies a given function to each element of the sequence.
search() Searches for matching sub-sequences.
search_n() Searches for matching sub-sequence of n consecutive elements.
 Algorithms ❖ 591❖

MUTATING ALGORITHMS
Table 25.4 – Mutating algorithms ( Header file <algorithm> )

Algorithm Description
copy() Makes a copy from start to end at a new location.
copy_backward() Makes a copy at new location in backward direction.
fill() Replaces elements of a sequence with copies of a value or object.
fill_n() Replaces the first n elements of sequence with a given value.
generate() Fills the sequence with values obtained by calling a function.
generate_n() The generate () is applied to first n elements of the sequence.
iter_swap() Swaps the values of elements at specified positions.
iota () The iota assigns a value to first element, value+1 to second element
and so on.
power() power (x, n) returns xn.
random_sample() Randomly copies the elements of a sequence. Any one elements
appears only once in output.
random_sample_n() Randomly copies n elements of a sequence.
random_shuffle () It rearranges the elements in random fashion.
remove() Removes elements having specified value in a sequence.
remove_copy() Copies sequence without the elements having specified value. The
original sequence is left intact.
remove_copy_if() Copies sequence elements that do not satisfy a given predicate.
remove_if() Removes the elements which satisfy a given predicate.
replace() Replaces the values of elements with new values.
replace_copy() Makes another sequence by replacing elements of a sequence.
replace_copy_if() Makes new sequence by replacing values of elements whose old
values satisfy a predicate.
replace_if() Replaces element values whose old values satisfy a predicate.
reverse() Reverses the order of elements of a sequence in a given range.
reverse_copy() Makes a new sequence with elements of a sequence reversed.
rotate() Rotates the elements of a sequence as if the two ends are
connected.
rotate_copy() Makes a new sequence by rotating the elements of a sequence.
Contd...
❖ 592 ❖ Programming with C++

Algorithm Description
swap() Swaps values of two objects of same type.

swap_ranges() Swaps elements in the specified ranges of two sequences.

transform() The elements of a sequence are transformed as a result of user

defined operation.

unique() Removes adjacent duplicate elements in a sorted sequence.

unique_copy() Makes a copy of sequence without adjacent duplicate elements.

SEARCH AND SORT ALGORITHMS

Table 25.5 – Search and sort algorithms (Header file <algorithm>)

Algorithm Description
binary_search() Carries out binary search for a value in a sorted sequence.
equal_range() It searches a sorted sequence for a range of elements equal to
a given value.
lower_bound() The function searches sorted sequence for the first position where
a given value may be inserted.
merge() Produces a sorted sequence by merging two sorted sequences.
inplace_merge() Merges two parts of same sequence.
is_sorted() The function returns true if the sequence is sorted in ascending
order.
nth_element() Sorting is carried out as much as it is necessary to place nth
element in its correct sorted position.
partial_sort() Sorts a specified number of elements from the beginning of the
sequence in ascending order.
partial_sort_copy() Makes a copy of a number of elements of a sequence and sorts
it.
partition() Partitions a sequence such that the elements that satisfy a
predicate are placed in the first part.
stable_partition() The function while partitioning maintains the relative order of
elements in both the parts.
sort() It sorts the sequence in ascending order.
stable_sort() In sorting the original order of equal elements is maintained.
upperbound() The function searches a sorted sequence for the last occurrence
of a given value.
 Algorithms ❖ 593❖

ALGORITHMS ON SETS
Table 25.6 – Algorithms on sets ( Header file <algorithm>)

Algorithm Description
includes() It returns true if all elements of second set are in the
first set. Both the sets should be sorted.
set_difference() Makes a set of elements of the first set which are not
in the second set.
set_intersection() Makes a set out of intersection of two given sets
(common elements of the two sets).
set_symmetric_difference() Produces a sequence of elements that are members of
either set but not of both.
set_union() Produces a set out of union of two sets.

NUMERIC ALGORITHMS
Table 25.7 – Algorithms for manipulation of numeric values (Header file <numeric>)

Algorithm Description
accumulate() Sums up the values of the elements of a segment of sequence
to the initial value.
adjacent_difference() Creates adjacent differences (S[n] – S[n – 1]) and loads these
on to another sequence.
inner_product() Returns the inner product of elements of two sequences of
same size.
partial_sum() Makes a sequence by partial sum of range of elements in a
sequence.

PERMUTATION ALGORITHMS
Table 25.8 – Algorithms for permutation (Header file <algorithm>)

Algorithm Description
next_permutation() Produces permutations of a sequence into next lexico-
graphically greater permutation.

prev_permutation() Produces permutations of a sequence into next permutation

which is lexicographically smaller sequence.
❖ 594 ❖ Programming with C++

ALGORITHMS ON HEAP
Table 25.9 – Algorithms on heap (Header file <algorithm>)

Algorithm Description
make_heap() Elements of a sequence are rearranged into a heap.
pop_heap() Removes the first element and rest are rearranged into heap.
push_heap () An element is added to heap.
sort_heap () Sorts the elements of a sequence into a heap.

25.3 ILLUSTRATIVE APPLICATIONS OF SOME ALGORITHMS

Simple typical applications of some of the algorithms are illustrated by the following programs
which are listed in alphabetical order.

ALGORITHM ACCUMULATE()

The syntax for accumulate() is of the following type

T accumulate (iterator start, iterator end, T initial_value);
The function sums up the values of element between the start and end iterators. The
sum is initialized by inititial_value. Program 25.1 illustrates it.

PROGRAM 25.1 – Illustrates accumulate() algorithm.

#include<iostream>
#include<numeric>
using namespace std;
int main()
{ int S[8] = {1,2,3,4,5,6,7,8};
int sum_four = accumulate( S , S+4, 0); //initial value is 0
cout<< “Sum of first 4 elements = ” << sum_four<<endl;
int sum_all = accumulate( S , S+8, 0);
cout << “Sum of all the element = ” << sum_all<<endl;
cout<<“Sum of all elements + 100 = ”<<accumulate(S,S+8,100)<< endl;
// initial value is 100.
cout<<“Sum of all elements –100 = ” <<accumulate(S,S+8,–100)<< endl;
// initial value is –100.
return 0 ;
}
 Algorithms ❖ 595❖
The expected output is given below.
Sum of first 4 elements = 10
Sum of all the element = 36
Sum of all elements + 100 = 136
Sum of all elements –100 = –64

ALGORITHM ADJACENT_DIFFERENCE()

Let there be a sequence ( a, b, c, d, e,). The function adjacent_difference produces the sequence
of adjacent differences (a, b – a, c – b, d – c, e – d) and loads it on another sequence.

PROGRAM 25.2 – Illustrates application of adjacent_difference() algorithm.

#include<iostream>
#include<numeric>
using namespace std;
int main()
{
int S[5] = { 5,20,40, 80, 160};
int A[5];
adjacent_difference( S, S+5,A); // A is another sequence.
for (int i =0; i<5;i++)
cout<< A[i]<<“ “;
cout <<“\n”;
return 0 ;
}

The output given below is self explanatory.

5 15 20 40 80

ALGORITHM ADJACENT_FIND ()
The syntax of the function is
iterator adjacent_find( iterator start, iterator end );
iterator adjacent_find( iterator start, iterator end,
binary_predsicate BP);
The function searches from start to end for two consecutive elements having identical
values. The function returns an iterator to the first of the two elements. If such an adjacent
pair is not found the return iterator points to the end of sequence. In the second version if
a predicate BP is provided to test then the predicate is used to determine equality of two
elements.
❖ 596 ❖ Programming with C++
PROGRAM 25.3 – Illustrates the application of adjacent_find() algorithm.

#include<iostream>
# include<algorithm> // includes header file <algorithm>
#include<vector> // includes header file <vector>
using namespace std;
int main()
{
vector<int> V;
int S[6] = { 10, 20, 40, 40, 50, 60};
for(int i =0; i < 6 ;i++)
V.push_back (S[i]); // constructing the vector.
int *Find = adjacent_find ( V.begin(), V.end() );
// Find is the iterator and *Find the value of element.
if ( Find == V.end())
{cout<<“ No match found”<<endl; }
else
cout<<“Found the match starting at ”<<*Find <<endl;
return 0 ;
}

Output is as under.
Found the match starting at 40

ALGORITHM BINARY_SEARCH()

The function binary_search () searches from start to end of a sequence for a specified value.
The sequence must be a sorted sequence. The function returns true (1) if the value is found
otherwise it returns false (0). See the following program for illustration.

PROGRAM 25.4 – Illustrates the function binary_search() algorithm.

#include<iostream>
# include<algorithm>
#include<vector>
using namespace std;
int main()
{
vector<int> V;
int S[7] = { 10,20,30 ,36, 44, 60, 70};
for(int i = 0; i< 7 ; i++)
 Algorithms ❖ 597❖
V.push_back (S[i]);
bool B = binary_search( V.begin(), V.end(), 36 );
// here 36 is the value to be searched.
if ( B!=0)
cout<<“The number “<<n<<“ is contained in the vector V.” <<endl;
else
cout<<n<< “The number is not in the vector V.”<<endl;
return 0 ;
}

The expected output is given below.

The number 36 is contained in the vector V.

ALGORITHM COPY()

The syntax is
iterator copy(iterator start, iterator end, iterator destination);
The function copies the elements from start to end to the destination specified as an
argument of the function. The return value of the function is iterator (or pointer) to the last
element copied.

PROGRAM 25.5 – Illustrates copy() algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{
int S[11] = { 10,20,30 ,36, 44, 60, 70, 80, 90,100,110};
int A[6];
copy( S+2 , S+8, A ); // A is the destination
cout<< “The elements of A are as below.\n”;
for(int i =0; i< 6; i++)
{cout << A [i] <<“ “;}
cout<<“\n”;
return 0 ;
}

The expected output is given below.

The elements of A are as below.
30 36 44 60 70 80
❖ 598 ❖ Programming with C++

The algorithm copies from element at index 2 to element at index 8. The S+8th element
is not included.

ALGORITHMS COPY_BACKWARD()

The function is similar to copy(), the difference is that it starts depositing elements from
back end.

PROGRAM 25.6 – Illustrates the application of copy_backward().

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{ int S1[10] = { 1,2,3,4,5,6,7,8,9,10};
int S2[10] = {0};
int S3[10] = {0};
cout << “S1 is : ”;
for (int i =0 ; i<10; i++)
cout<<S1[i]<<“ ”;

cout <<“\nS2 is : ”;
copy(S1, S1+10, S2);

for(int n =0; n<10; n++)

cout <<S2[n]<<“ ” ;

copy_backward( S1, S1 + 6 , S3+10 );

cout<<“\nS3 is : ”;
for ( int j = 0; j< 10 ; j++)
cout<<S3[j]<<“ ”;
cout<<“\n”;
return 0;
}

The expected output is as below. The output is self explanatory.

S1 is : 1 2 3 4 5 6 7 8 9 10
S2 is : 1 2 3 4 5 6 7 8 9 10
S3 is : 0 0 0 0 1 2 3 4 5 6

ALGORITHM COUNT()

The function returns the number of elements in the sequence that matches a specified value.
The following program illustrates the application of the function.
 Algorithms ❖ 599❖
PROGRAM 25.7 – Illustrates the algorithm count().

#include<iostream>
#include<algorithm>
#include<vector>
using namespace std;
int main()
{
vector<int> V;
int S[8] = { 10,20,24 ,36, 24, 24, 70};
for(int i =0; i < 6 ;i++)
V.push_back (S[i]);
int n = 24;
int m = count( V.begin(), V.end(), n );
cout<<“The number of elements with value “<<n<<“ in V are = “<< m <<endl;
return 0 ; }

The expected output is given below.

The number of elements with value 24 in V are = 3

ALGORITHM COUNT_IF()

The application of this algorithm has already been illustrated in Chapter 21 on STL. Below
is another application.

PROGRAM 25.8 – Illustrates application of count_if()algorithm.

#include<iostream>
#include<algorithm>
#include<functional> //This is needed for predicates.
using namespace std;
int main()
{
int S[ ] = { 5,6,8,7,8,3,8,10,8, 12};
int n= count_if(S, S+10, bind2nd(greater<int>(),7));
//n is count of elements greater than 7 in S
cout<< “Number of elements of S >7 are = “ <<n <<endl;
int m = count_if(S, S+10, bind2nd(less<int>(),10));
// m is count of elements less than 10
cout<< “Number of elements of S < 10 are = “ <<m <<endl;
return 0;
}
❖ 600 ❖ Programming with C++

The expected output is given below. For more details on predicates and binder functions
see Chapter 21.
Number of elements of S >7 are = 6
Number of elements of S < 10 are = 8

ALGORITHM EQUAL()

The function equal() returns true if the elements in the two ranges are same, otherwise it
returns false. The function takes three arguments, the first two are the start and end of the
first sequence and the third is the start of second sequence.

PROGRAM 25.9 – Illustrates equal()algorithm.

#include<iostream>
#include<algorithm>
#include<vector>
using namespace std;
int main()
{
vector<int> V1;
int S1[6] = { 10,13,25 ,36, 25, 50};
for(int i =0; i < 6 ;i++)
V1.push_back (S1[i]);
vector<int> V2;
int S2[6] = { 10,13,25,36,25,25};
for(int j =0; j < 6 ;j++)
V2.push_back (S2[j]);
int S3 [] = {11,12,13, 4,5,11,12,13};
vector<int> V3;
for ( int k =0; k<8 ; k++)
V3.push_back ( S3[k]);
if (equal( V3.begin(), V3.begin()+3, V3.begin()+5));
cout << “The two subsets of three values are equal.”<<endl;
if(equal( V1.begin(), V1.end(), V2.begin ()))
cout<<“The two vectors are equal.” <<endl;
else
cout<<“The two vectors are not equal.”<<endl;
return 0 ;
}
 Algorithms ❖ 601❖

The expected output is given below.

The two subsets of three values are equal.
The two vectors are not equal.
In the above program the function equal is tested on elements of two vectors (second
line of output) as well on two segments of the same vector. See the first line of the output.

ALGORITHMS EQUAL_RANGE ()
The function equal_range returns the range of elements which are equal to a specified value
in a sorted sequence. The following program illustrates the same.

PROGRAM 25.10 – Illustrates equal_range() algorithm.

#include<iostream>
#include<algorithm>
#include<vector>
using namespace std;
int main()
{ vector<int> V;
int S[8] = { 10,20, 24 ,24, 24, 36, 70};
for(int i =0; i < 6 ;i++)
V.push_back (S[i]);
int n = 24;
pair <vector<int>::iterator, vector<int>::iterator> Range;
Range = equal_range ( V.begin(), V.end(), n );
cout<<“The equal range starts after ”<<*( Range.first –1) <<“ and ends before”
<<*(Range.second )<<endl;
return 0 ;
}

The expected output is given below.

The equal range starts after 20 and ends before 36

ALGORITHMS FILL() AND FILL_N ()

The function assigns the specified value to all the elements from a start to an end. The function
fill_n assigns values to the first n elements of the sequence.

PROGRAM 25.11 – Illustrates the fill() algorithm and fill_n() algorithm.

#include<iostream>
#include<algorithm>
#include<vector>
using namespace std;
int main()
❖ 602 ❖ Programming with C++
{
vector<int> V;
int S[9] = { 15,20, 30 , 40, 50, 60, 70, 80, 90};
for(int i =0; i < 9 ;i++)
V.push_back (S[i]);
int m =5;
fill_n ( V.begin(), V.size()/3 , m );
for (int k =0; k<V.size(); k++)
cout << V[k] << “ ” ;
cout<< “\n”;
int n = 10;
fill ( V.begin(), V.end(), n );
for (int j =0; j<V.size(); j++)
cout << V[j] << “ ” ;
cout<< “\n”;
return 0 ; }

The expected output is given below.

5 5 5 40 50 60 70 80 90
10 10 10 10 10 10 10 10 10
In the first line of output the value 5 replaces the previous values of first three elements
of vector V. In the second line of output, all the elements of vector V are filled with a value
10 by function fill().The previous values are removed.

ALGORITHM FIND()

The algorithm find() return iterator to the element that matches a specified value. If the
value is not found, it returns iterator to end of sequence. The following program illustrates
the same.

PROGRAM 25.12 – Illustrates application of find()algorithm.

#include<iostream>
#include<algorithm>
#include<vector>
using namespace std;
int main()
{ vector<int> V;
int S[7] = { 10,20, 24 ,30, 35, 36, 70};
for(int i =0; i < 7 ;i++)
V.push_back (S[i]);
int n = 24;
vector <int> ::iterator iter;
 Algorithms ❖ 603❖
iter = find(V.begin(), V.end(), n);
if (iter != V.end())
cout << “The number “<<n<<“ is an element of vector.”<<endl;
else
cout<< “The number “ << n<<“ is not an element of vector.\n”;
return 0 ;
}

The expected output is given below.

The number 24 is an element of vector.

ALGORITHM FIND_END ()
The function searches for a sequence of elements in another sequence. If the sequence is found,
the algorithm returns iterator to first element of the last sequence found. Otherwise it returns
iterator to end of sequence.

PROGRAM 25.13 – Illustrates the algorithm find_end() algorithm.

#include<iostream>
#include<algorithm>
#include<vector>
using namespace std;
int main()
{ vector<int> V;
int S[12] = { 10,20,25 ,26, 27, 36, 20,25 ,26,27, 70, 90};
for(int i =0; i < 12 ;i++)
V.push_back (S[i]);

int Find_Seq1 [] = {20, 25,26,27};

vector<int>:: iterator iter;
iter = find_end( V.begin(), V.end(), Find_Seq1, Find_Seq1+3);
if ( iter != V.end())

cout<<“The last Find_Seq1 starts at ”<< *iter <<endl;

else
cout<< “There is no Find_Seq1 in the vector.”<<endl;

int Find_Seq2 [] = {20, 4,26 };

iter = find_end( V.begin(), V.end(), Find_Seq2, Find_Seq2+2);
if ( iter != V.end())
cout<<“The last Find_Seq2 starts at “<< *iter <<endl;
else
cout<< “There is no Find_Seq2 in the vector.”<<endl;
return 0 ;
}
❖ 604 ❖ Programming with C++

The expected output is given below. The output is self explanatory.

The last Find_Seq1 starts at 20
There is no Find_Seq2 in the vector.

ALGORITHM FIND_FIRST_OF()

The function finds the first occurrence of an element of a sequence in a sequence.

PROGRAM 25.14 – Illustrates find_first_of()algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{
int * Find;
int S[9] = { 1,6,2,3,4,5,6,7,8};
int S2[] = { 5,6,7};
Find = find_first_of( S, S+9, S2, S2+3);
if(Find != S+9) //if no match is found the pointer points to end
cout<<“Found the match :” << *Find<<endl;
return 0 ;
}

The expected output is given below.

Found the match :6

ALGORITHM FIND_IF()

It searches for a value subject to satisfaction of a predicate. In the following program the
search is for element greater than 7. Since all the elements have to be compared with the
number 7, the number 7 is bound to the function greater<int>() by the binder function
bind2nd. See Chapter 21 for details on function objects, predicates and binders.

PROGRAM 25.15 – Illustrates find_if()algorithm.

#include<iostream>
#include<algorithm>
#include<functional>
using namespace std;
int main()
{
int S[ ] = { 5,6,3,7,8,3,8,10,8, 12};
int* n = find_if(S, S+10, bind2nd(greater<int>(),7));
 Algorithms ❖ 605❖

cout<< “*n = ”<< *n<<endl;

cout << “The position of element = ”<<n+1–S <<“th”<<endl;

return 0;
}

The expected output is given below.

*n = 8
The position of element = 5th

ALGORITHM FOR_EACH()

The algorithm applies the specified function to each element of a sequence.

PROGRAM 25.16 – Illustrates the application of for_each() algorithm

#include<iostream>
#include<algorithm>
#include<functional>
using namespace std;
void Multiplies ( int &x)
{cout<< 10*x <<“ ”;}
int main()
{ int S1[ ] = { 5,6,8,7,4,3,8,10,11, 12};
for_each ( S1, S1+10, Multiplies );
cout<<“\n”;
return 0;
}

The expected output is given below.

50 60 80 70 40 30 80 100 110 120
Clearly the function multiplies each element by 10 and displays it.

ALGORITHMS GENERATE() AND GENERATE_N()

The two functions are used to generate values through the operation of a function. In the
following program the generate() is used to generate 8 random numbers each of 2 digits.
The second function generate_n() is used to generate a specified number of random
numbers.

PROGRAM 25.17 – Illustrates generate() and generate_n() algorithms.

#include<iostream>
#include<algorithm>
❖ 606 ❖ Programming with C++
using namespace std;
void Display (int x)
{cout<<x<< “ “;}
int main()
{ int S[8] = { 0 };
generate (S, S+8, rand );
for ( int i =0; i<8; i++)
cout<< S[i]%100<<“ “;
cout<<“\n”;
int B [6] = {0};
generate_n( B , 6, rand);
for(int j =0; j<6; j++)
cout<< B[j] <<“ “;
return 0 ;
}

The expected output is given below.

41 67 34 0 69 24 78 58
26962 24464 5705 28145 23281 16827

ALGORITHM INCLUDES()

The function returns a Boolean value if the elements of a set or specified segment of a set is
included in another set. In the following program includes() is used to test if a segment
of a set is included in another set.

PROGRAM 25.18 – Illustrates the includes()algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{ int S[8] = { 5,6,7,8,9,11,12,13 };
int A [9] = { 5,6,7,8, 21,22,23,24,25};
bool B ;
B = includes (S, S+8, A, A+4); //test if A to A+4 are in S
if (B)
cout << “The Sequence S includes A[0] to A[3]” << endl;
else
cout << “The Sequence S does not includes A[0] to A[3]” << endl;
return 0 ;
}
 Algorithms ❖ 607❖

The expected output is given below.

The Sequence S includes A[0] to A[3]

ALGORITHM INNER_PRODUCT()

This algorithm is part of header file <numeric>. So this header file must be included in the
program. The algorithm inner_product() does the inner vector product of two vectors.
For example if the two vectors have elements (2, 4, 5) and (1, 3, 6), the inner product =
2×1 + 4×3 + 5×6 = 44.

PROGRAM 25.19 – Illustrates inner_product() algorithm.

#include<iostream>
#include<numeric>
using namespace std;
int main()
{ int S[7] = { 6,4,8,9,10,20,10};
int A[7] = { 2,2,2,2,2,2,2};
int n = 20;
// the value of n initializes the inner_product.
int Innerproduct = inner_product ( S , S+7, A, n );
cout<< “Innerproduct of elements S and A plus “<<n<<“ = “ <<
Innerproduct<<endl;
return 0 ;
}

The expected output is given below.

Innerproduct of elements S and A plus 20 = 154

ALGORITHM INPLACE_MERGE()

The function merges two parts of the same sequence and results in a sorted sequence.

PROGRAM 25.20 – Illustrates inplace_merge() algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{
int S[8] = { 2,4,5,1,3,6,7,9 };
int A [8] ={ 12,13,14,15, 5,6,7, 24} ;
inplace_merge ( S, S+3, S+8);
inplace_merge ( A, A+4, A+8);
❖ 608 ❖ Programming with C++
for(int i =0; i<8; i++)
cout << S[i] <<“ ”;
cout<<“\n”;
for ( int j =0; j<8; j++)
cout<< A [j] <<“ ”;
return 0 ;
}

The expected output is given below.

1 2 3 4 5 6 7 9
5 6 7 12 13 14 15 24

ALGORITHM ITER_SWAPS()

The function swaps the elements of two sequences at specified locations.

PROGRAM 25.21 – Illustrates iter_swap() algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{ int S[8] = { 2,4,5,1,3,6,7,9 };
int A [8] ={ 12,13,14,15, 5,6,7, 24} ;
iter_swap ( S+4, A+3);
for ( int i =0; i<8; i++)
cout << S[i] <<“ “;
cout<<“\n”;
for ( int j =0; j<8; j++)
cout<< A [j] <<“ “;
return 0 ;
}

The expected output is given below. The value 15 of A has been exchanged with 3 of S.
2 4 5 1 15 6 7 9
12 13 14 3 5 6 7 24

ALGORITHM LEXICOGRAPHICAL_COMPARE()

The function compares two sets lexicographically, i.e. dictionary style. For such a function it
is best to take strings for comparison. In the following program the string “AA” is compared
with “ZZ” lexicographically. Evidently the first string is smaller than the second, hence, the
output is 1 (true). If the second is compared with the first the result is 0. In the third case
 Algorithms ❖ 609❖

a string is compared with itself. The result is 0 because it cannot precede itself. In the third
case a string “compute” is compared with string “computer”. The output is obviously 1.

PROGRAM 25.22 – Illustrates lexicographical_compare() algorithm.

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{
char S1[ ] = “AA”;
char S2 [ ] =”ZZ” ;
char B[] = “compute”;
char C[] =”computer”;
bool b1 =lexicographical_compare ( S1, S1+1, S2, S2+1);
cout<<“b1 = ” << b1<<endl;
bool b2 =lexicographical_compare (S2, S2+1 ,S1, S1+1 );
cout<< “b2 = ” << b2<<endl;
bool b3 = lexicographical_compare ( S1, S1+1, S1, S1+1);
cout<<“b3 = ” <<b3<<endl;
bool b4 = lexicographical_compare ( B, B+7, C, C+8);
cout<<“b4 = ” << b4<<endl;
return 0 ;
}

The expected output is as below.

b1 = 1
b2 = 0
b3 = 0
b4 = 1

THE ALGORITHMS LOWER_BOUND() AND UPPER_BOUND()

The functions are applied to sorted sequences. The lower_bound gives the position of first
occurrence of a value in the sequence. The upper_bound gives the position of element
next to last occurrence of a value in the sequence. The following program illustrates both the
functions. First the function lower_bound is applied to an array of integers for a value 20. It
is at index value 6. The upper_bound gives a value 9. That is index value of 22 in the array.

PROGRAM 25.23 – Illustrates application of lower_bound() and upper_bound().

#include<iostream>
# include<algorithm>
❖ 610 ❖ Programming with C++
using namespace std;
int main()
{ int S1[ ] = { 11, 12, 13 ,16, 16 , 19,20, 20, 20 , 22, 24 };
int* ptr1 = lower_bound( S1, S1+11 ,20);
cout <<“lower_bound for 20 = ” <<ptr1 —> S1 <<endl;

int* ptr2 = lower_bound( S1, S1+11 ,16);

cout <<“lower_bound for 16 = ”<< ptr2 —> S1 <<endl;
int *uptr1 = upper_bound ( S1, S1+11, 16);

cout <<“upper_bound of 16 is = ” << uptr1 —> S1 <<endl;

int* Uptr2 = upper_bound ( S1, S1+11, 20);
cout << “upper_bound of 20 = ” <<Uptr2 —> S1 <<endl;
return 0 ;
}

The expected output is given below.

lower_bound for 20 = 6
lower_bound for 16 = 3
upper_bound of 16 is = 5
upper_bound of 20 = 9

ALGORITHM MAKE_HEAP()

The function turns a given sequence of elements into heap.

PROGRAM 25.24 – Illustrates make_heap() algorithm
#include<iostream>
#include<algorithm>
using namespace std;

int main()
{
int S[13] = {10, 12, 22,24 ,34,51,71,13,16 ,177 ,106 ,6,7};
make_heap( S,S+13);

for ( int i =0; i<13; i++)

cout << S[i] <<“ “;
cout<<“\n”;

return 0 ;
}

The expected output is given below.

177 106 71 24 34 51 22 13 16 10 12 6 7
 Algorithms ❖ 611❖

ALGORITHM MAX_ELEMENT()

The function returns position of element with maximum value.

PROGRAM 25.25 – Illustrates application of max_element() algorithm.

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{ int S1[ ] = { 11, 12, 113 ,216 , 19,20 };
cout<< *max_element ( S1, S1+6) <<endl;
const int * ptr= max_element ( S1, S1+6);
// value of max_element
cout<< ptr – S1 <<endl; // position of max element
return 0 ;
}

The following output is obvious.

216
3

ALGORITHM MERGE()

The algorithm merges two sorted sequences to create a new sorted sequence.

PROGRAM 25.26 – Illustrates merge()algorithm

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{ int S1[ ] = {10, 12, 40 };
int S2[] = {34,51,71 ,106 ,177 };
int A[8];
int B[6];
merge ( S1, S1+3, S2, S2+5 , A);
for ( int i =0; i<8 ;i++)
cout<< A[i]<<“ “;
cout<<“\n”;

merge ( S2, S2+4,S1, S1+2, B);

for ( int k =0; k<6 ;k++)
cout<< B[k]<<“ “;
cout<<“\n”;
return 0 ;
}
❖ 612 ❖ Programming with C++

The expected output is given below.

10 12 34 40 51 71 106 177
10 12 34 51 71 106

ALGORITHM MIN_ELEMENT()

The function returns iterator to the element with minimum value.

PROGRAM 25.27 – Illustrates min_element()algorithm.

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{
int S1[ ] = { 19,20, 11, 12, 113 ,216};
cout<< *min_element ( S1, S1+6) <<endl; // value of min. element
const int * ptr= min_element ( S1, S1+6);
cout<< ptr —> S1 <<endl; // gives index value of min. element
return 0 ;
}

The expected output is given below.

11
2

ALGORITHM MISMATCH()

The function returns a pair of iterators which give position of mismatch between two
sequences. The following program illustrates the same.

PROGRAM 25.28 – Illustrates mismatch() algorithm

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{
char S1[] = “Computer, Computar, Computor”;
char S2[] = “Computer, Computer, Computer”;
int m = strlen(S1);
pair<char*, char*> p = mismatch( S1, S1+m, S2);
char* S1ptr= p.first;
 Algorithms ❖ 613❖
char* S2ptr= p.second;
cout<<“length = “<< m <<endl;
cout<< “First location of mismatch in string S1 is at “<<S1ptr - S1<<endl;
The pointers S1 and S2 are subracted to get index values.
cout<< “First location of mismatch in string S2 is at “<<S2ptr - S2<<endl;
return 0 ;
}

The expected output is given below. The result is obvious.

length = 28
First location of mismatch in string S1 is at 16
First location of mismatch in string S2 is at 16

ALGORITHM NEXT_PERMUTATION()

Gives another permutations of a sequence

PROGRAM 25.29 – Illustrates next_permutation()algorithm.

#include <iostream>
#include<algorithm>
using namespace std;
void main()
{
char A[5] = “XYZ”;
for( int k =0; k<6;k++)
{ next_permutation (A, A+3);
cout<<A<<“ “;}
cout<<“\n”;
}

The expected result is given below.

XZY YXZ YZX ZXY ZYX XYZ

ALGORITHM NTH_ELEMENT()

The function partitions the sequence into two and rearranges the sequence about the specified
middle element. The elements with values less than that of the specified element are sorted
and put before the element and those with values more than the specified value are put after
the specified element. The following program illustrates the function.
❖ 614 ❖ Programming with C++
PROGRAM 25.30 – Illustrates nth_element()algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{
int S[10] = {10, 12, 22,24 ,34,51,171,13,16 ,170};
nth_element(S, S+4, S+10);
for ( int i =0; i<10; i++)
cout << S[i] <<“ “;
cout<<“\n”;
return 0 ;
}

The expected output is given below.

10 12 13 16 22 24 34 51 170 171

ALGORITHM PARTIAL_SORT()

The function sorts the sequence up to a specified position, the remaining elements are placed
after the specified position. See the following program for illustration.

PROGRAM 25.31 – Illustrates partial_sort()algorithm.

#include<iostream>
#include<algorithm>
using namespace std;
int main()
{
int S[12] = {10, 24,12, 22 ,34,51,171,13,16, 176, 56, 42};
partial_sort(S, S+6, S+12); // sort up to s+6
for ( int i =0; i<12; i++)
cout << S[i] <<“ “;
cout<<“\n”;
return 0 ; }

The expected output is given below.

10 12 13 16 22 24 171 51 34 176 56 42

ALGORITHM PARTIAL_SORT_COPY

The function makes a copy of the partially sorted elements. The original sequence remains intact.
 Algorithms ❖ 615❖
PROGRAM 25.32 – Illustrates partial_sort_copy()algorithm.
#include<iostream>
# include<algorithm>
using namespace std;
int main()
{ int S[12] = {10,24,12,22,34,51,171,115,20,176, 56, 42 };
int A [6];

partial_sort_copy (S, S+12, A, A+6);

cout<<“ S = ”;
for ( int i =0; i<12; i++)
cout << S[i] <<“ ”;

cout<<“\nA = ”;
for (int k =0; k<6; k++)
cout << A[k] <<“ ”;
cout<<“\n”;
return 0 ;
}

The expected output is given below.

S = 10 24 12 22 34 51 171 115 20 176 56 42
A = 10 12 20 22 24 34

ALGORITHM PARTIAL_SUM ()
The function creates a new sequence by adding the elements of existing sequence up to the
element position. Thus if a, b, c, d are the elements of a sequence, the function partial sum
makes the sequence a, a + b, a + b + c, a + b + c + d and loads it to the new sequence.
The following program illustrates the same.

PROGRAM 25.33 – Illustrates partial_sum() algorithm.

#include<iostream>
#include<numeric>
using namespace std;
int main()
{
int S[8] = {2,3,4,5,6,7,8,9 };
int A [8];
cout<< “S = ”;
for ( int i =0; i<8; i++)

cout << S[i] <<“ ”;

partial_sum (S, S+8, A);
cout<<“\nA = ”;
❖ 616 ❖ Programming with C++
for ( int k =0; k<8; k++)
cout << A[k] <<“ ”;
cout<<“\n”;
return 0 ;
}

The expected output is given below.

S=23456789
A = 2 5 9 14 20 27 35 44

ALGORITHM PARTITION()

The function partitions a sequence such that the elements which satisfy a predicate are put in
the first part followed by the elements which do not satisfy the predicate. See the following
program for illustration.

PROGRAM 25.34 – Illustrates the function partition().

#include<iostream>
#include<algorithm>

using namespace std;

bool Even(int m ) // definition of function for even number

{return !(m%2) ? true: false ;}

int main()
{
int S[ ] = { 5,6,8,7,4,3,8,10,11, 12};
partition(S, S+10, Even );
//partition and put even numbers first.
for ( int i =0; i<10 ; i++)
cout<< S[i]<< “ “;
return 0;
}

The expected output is given below.

12 6 8 10 4 8 3 7 11 5

ALGORITHM PREV_PERMUTATION()

It is inverse of next_permutation. The following program illustrates it.

PROGRAM 25.35 – Illustrates prev_permutation()algorithm.

#include <iostream>
#include<algorithm>
using namespace std;
 Algorithms ❖ 617❖
void main()
{
char A[4] = “XYZ”;
for( int k =0; k<6;k++)
{ next_permutation (A, A+3);
cout<<A<<“ “;}
cout<<“\n”;
for( int i =0; i<6;i++)
{prev_permutation (A, A+3);
cout<<A<<“ “;}
}

The output is given below.

XZY YXZ YZX ZXY ZYX XYZ
ZYX ZXY YZX YXZ XZY XYZ

ALGORITHM RANDOM_SHUFFLE()

The function performs a random shuffle of elements of a sequence.

PROGRAM 25.36 – Illustrates random_shuffle() algorithm.

#include <iostream>
#include<algorithm>
using namespace std;
void main()
{ char A[7] = “ABCDEF”;
for( int k =0; k<5;k++)
{random_shuffle(A, A+6);
cout<<A<<“ “;}
cout<<“\n”;
}

The output of the program is given below.

EDACFB CEFABD EDCBFA DABFEC BECADF

THE ALGORITHMS REMOVE() AND REMOVE_IF()

The function removes the specified element from the sequence. In case of remove_if the
element is removed if it satisfies a given predicate. See following program for illustration.

PROGRAM 25.37 – Illustrates remove() and remove_if() algorithms.

#include<iostream>
#include<algorithm>
❖ 618 ❖ Programming with C++
#include<functional>
#include <vector>
using namespace std;
vector <int> V;
int main()
{
int S[ ] = { 5,6,8,7,8,3,8,10,8,12};
for (int i =0 ; i<10;i++)
V.push_back (S[i]); //construct a vector V with elements of S
cout<< “V = ”;
vector<int>:: iterator iter; // declaration of iterator
for (iter= V.begin(); iter <V.end(); iter++)
cout<< *iter<<“ ”; // Display elements of V
cout<<endl;
int Count = count(V.begin(), V.end(), 8); // count number of 8s
cout<<“Number of digit 8 = ”<<Count<<endl;
remove ( V.begin(), V.end() , 8 ); // remove the digit 8s
for ( int j = 0; j<Count; j++)
V.pop_back( ) ;
cout<< “ Now V = ”;
for (iter= V.begin(); iter <V.end(); iter++)
cout<< *iter<<“ ”; // display elements after removing 8s
cout<<endl;
Count = count_if(V.begin(), V.end(), bind2nd(less<int>(),6) );
cout<<“ Number of elements less than 6 = ”<<Count;
remove_if(V.begin(), V.end(), bind2nd(less<int>(),6) );
// remove the number if less than 6
for ( int k = 0; k<Count; k++)
V.pop_back( ) ;
cout<< “\nThe new vector after two removals is as below.”<<endl;
cout<<“V = ”;
for (iter= V.begin(); iter <V.end(); iter++)
cout<< *iter<<“ ”;
cout<<endl;
return 0;
}

The expected output is as below.

V = 5 6 8 7 8 3 8 10 8 12
Number of digit 8 = 4
Now V = 5 6 7 3 10 12
Number of elements less than 6 = 2
 Algorithms ❖ 619❖
The new vector after two removals is as below.
V = 6 7 10 12

THE ALGORITHMS REPLACE()AND REPLACE_IF()

The following program illustrates the application of the two functions on a vector.

PROGRAM 25.38 – Illustrates the replace() and replace_if()algorithms.

#include<iostream>
#include<algorithm>
#include<functional>
#include <vector>
using namespace std;
vector <int> V;
int main()
{
int S[ ] = { 5,6,8,7,8,3,8,10,8, 12};
for (int i =0 ; i<10;i++)
V.push_back (S[i]);
vector<int>:: iterator iter;
for (iter= V.begin(); iter <V.end(); iter++)
cout<< *iter<<“ “;
cout<<endl;
replace ( V.begin(), V.end() , 8 , 9 ); // replace 8 by 9
for (iter= V.begin(); iter <V.end(); iter++)
cout<< *iter<<“ “;
cout<<endl;
replace_if(V.begin(), V.end(), bind2nd(less<int>(),9), 5);
// replace those less than 9 by 5
cout<< “The new vector after replacement is as below.”<<endl;
for (iter= V.begin(); iter <V.end(); iter++)
cout<< *iter<<“ “;
return 0;
}

The expected output is as under.

5 6 8 7 8 3 8 10 8 12
5 6 9 7 9 3 9 10 9 12
The new vector after replacement is as below.
5 5 9 5 9 5 9 10 9 12
❖ 620 ❖ Programming with C++

The output shows that elements with value 8 has been replaced by value 9 and then
those less than 9 are replaced by 5.

ALGORITHM REVERSE()

It reverses the order of elements in the sequence. See the following program for illustration.

PROGRAM 25.39 – Illustrates reverse()algorithm.

#include <iostream>
#include<algorithm>
using namespace std;
void main()
{
char A [ ] = “ABCDEFGH”;
int n = strlen(A);
reverse(A, A+4);
cout<< A<<endl;
}

The expected output is given below.

DCBAEFGH

ALGORITHM ROTATE()

The function rotates the elements as if it is endless chain.

PROGRAM 25.40 – Illustrates rotate()algorithm.

#include<iostream>
# include<algorithm>
using namespace std;
void main()
{
int S1[ ] = { 11, 12, 13 ,15 ,16,17, 18,19 ,20};
cout<< “S1= ”;
rotate ( S1, S1+4,S1+9);
for ( int j =0; j<9; j++)
cout << S1[j]<<“ “;
}

The expected output is given below.

S1= 16 17 18 19 20 11 12 13 15
 Algorithms ❖ 621❖

THE ALGORITHMS DEALING WITH SETS

set_difference(),set_intersection(),set_symmetric_difference()and
set_union ()
These algorithms concerning sets are illustrated below.

PROGRAM 25.41 – Illustrates set_difference(), set_intersection(), set_

symmetric_difference()and set_union() algorithms.

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{
char Set1[ ] = { “ACDGHIJKTUVZ” };
char Set2[] = { “BCDKMNPUV”};
char Setx[12];
char Setxy[15];
char SetA[20];
char SetU[25];
cout<<“Set1 = ”<<Set1<<endl;
cout<<“Set2 = ”<< Set2<<endl;
char* Sety = set_difference(Set1, Set1+12, Set2, Set2+9, Setx);
*Sety =0;
cout<<“ set_difference = ”<<Setx <<endl;
char* Setz = set_intersection(Set1, Set1+12, Set2, Set2+9,Setxy);
*Setz =0;
cout <<“set_intersection = ”<<Setxy<<endl;
char* Setm = set_symmetric_difference(Set1, Set1+12, Set2,Set2+9, SetA);
*Setm = 0;
cout <<“Set_symmetric_difference = ”<< SetA<<endl;
char* Setn = set_union(Set1, Set1+12 , Set2, Set2+9 , SetU);
*Setn =0;
cout << “Union of Set1 and Set2 = ”<<SetU <<endl;
return 0 ;
}

The expected output is given below.

Set1 = ACDGHIJKTUVZ
Set2 = BCDKMNPUV
set_difference = AGHIJTZ
❖ 622 ❖ Programming with C++
set_intersection = CDKUV
Set_symmetric_difference = ABGHIJMNPTZ
Union of Set1 and Set2 = ABCDGHIJKMNPTUVZ

ALGORITHM SORT()

Sorts a sequence.

PROGRAM 25.42 – Illustrates sort()algorithm.

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{
int S1[ ] = { 11, 12, 19 ,13 ,16 ,20,24 ,22};
sort (S1, S1+8);
for ( int i =0; i<8;i++)
cout << S1[i]<<“ ”;
cout<<“\n”;
return 0 ;
}

The expected output is given below.

11 12 13 16 19 20 22 24

ALGORITHM SWAP_RANGES()

The function exchanges the elements in a range with the elements of an another range of
same size.
PROGRAM 25.43 – Illustrates swap_ranges()algorithm.
#include<iostream>
#include<algorithm>
using namespace std;
void main()
{
int S1[ ] = { 11, 12, 19 ,13 ,16 ,20, 24 ,22};
int S2[] = { 1, 2, 3, 4, 5, 6, 7, 8};

cout<< “S1= ”;
swap_ranges ( S1, S1+4, S2+4);
for ( int j =0; j<8; j++)
cout << S1[j]<<“ ”;
cout<< “\nS2= ” ;
for ( int k =0; k<8; k++)
cout << S2[k]<<“ ”;
}
 Algorithms ❖ 623❖
The expected output is given below.
S1= 5 6 7 8 16 20 24 22
S2= 1 2 3 4 11 12 19 13

ALGORITHM UNIQUE(), REVERSE() AND REMOVE()

The unique () removes adjacent duplicate elements in a sorted sequence. The remove () removes
the elements having specified value in a sequence. The reverse () reverses the order of elements
of a given range.

PROGRAM 25.44 – Illustrates unique(), reverse() and remove()algorithms.

#include<iostream>
#include<list>
using namespace std;
list<char> L1 ;
void main()
{ for (int i=0; i<6;i++)
{L1.push_back(65 +i );
L1.push_back(65 + i );
}
cout<<“The original lists is as below.”<<endl;
list<char>::iterator T1;
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
cout<<endl;
L1.unique();
cout<<“ After removal of duplicate the list is:”<<endl;
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ “ <<*T1 ;
L1.reverse();
cout<<“\n After reversing the list it is \n”;
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
cout<<“\nAfter removal of F The list becomes \n”;
L1.remove (‘F’);
for ( T1 = L1.begin() ; T1!=L1.end() ; T1++ )
cout <<“ ” <<*T1 ;
cout << endl;
}

The output is given below.

The original lists is as below.
AABBCCDDEEFF
After removal of duplicate the list is:
ABCDEF
After reversing the list it is
FEDCBA
After removal of F The list becomes
EDCBA
❖ 624 ❖ Programming with C++

ALGORITHM UPPER_BOUND()

PROGRAM 25.45 – Illustrates upper_bound() algorithm.

#include<iostream>
# include<algorithm>
using namespace std;
int main()
{
int S1[ ] = { 11, 121, 19 ,122 ,260 ,203,240 ,22};
int* ptr = upper_bound (S1, S1+8, 122);
cout<< “The upper bound is ”<< *ptr<< endl;
cout<< “Its subscript is ”<< ptr - S1 << “ in the array.”<<endl;
return 0 ;
}

The expected output is given below.

The upper bound is 260
Its subscript is 4 in the array.

EXERCISES
1. What is an algorithm?
2. Which header file is included for using the alogrithm accumulate()?
3. What is the action of function adjacent_difference() on elements of a container?
4. Explain the action of algorithm adjacent_find().
5. Explain the mechanism of binary_search.
6. What is a predicate?
7. How would you use the algorithm find_if(), if it is required to separate the even and
odd elements of an int array?
8. Explain the difference between the algorithms count() and count_if().
9. Explain the actions of algorithm upper_bound() and lower_bound().
10. What does the algorithm equal_range() do?
11. Explain the difference between fill() and fill_n() algorithms.
12. Explain the action of algorithm find_end().
13. Explain the difference between the algorithms remove() and the algorithm remove_if().
14. Make a small program to explain the action of algorithm remove_if().
15. Explain the action of algorithms swap() and swap_ranges().
16. Make a small program to illustrate the action of algorithm rotate().
17. Make a small program to illustrate the action of algorithm for_each().
18. Explain the action of algorithm partition() on a sequence.
19. Explain the action of following algorithms on sets.
(i) set_difference()
(ii) set_intersection()
(iii) set_symmetric_difference()
(iv) set_union()
20. Make a program to illustrate set_union() and set_difference().

❍❍❍
 1. John R. Hubbard; Programming with C++ (Second Edition); McGraw-Hill, International
Editions, 2000.
2. Deitel H. M., Deital P. J.; C++ How to Program, Fifth Edition; Prentice-Hall of India
Pvt. Ltd, New Delhi, 2005.
3. Venugopal K. R., Rajkumar Buuya and Ravishankar T.; Mastering C++; Tata McGraw-
Hill Publishing Company Ltd., New Delhi, 2006.
4. Balagurusamy E.; Object Oriented Programming; Third Edition; Tata McGraw-Hill
Publishing Company Ltd., New Delhi, reprint 2006.
5. Jesse Liberty and Jim Keogh; C++: An Introduction to Programming; Prentice-Hall of India,
New Delhi, 1999.
6. Graham M. Seed; An Introduction to Object Oriented Programming in C++; 2nd Edition;
Springer, Berlin, 2001.
7. Daniel W. Lewis; Fundamentals of Embedded Software; Prentice-Hall of India Pvt. Ltd.,
New Delhi, 2003.
8. Gary J. Bronson; A First Book of ANSI C; (Brook/cole) Thomson Asia Pte Ltd., Singapore,
2001.
9. A. N. Kamthane; Object-Oriented Programming with ANSI & Turbo C++; Pearson Education
(Singapore) Pte. Ltd., 2003.
10. Bjarne Stroustrup; The C++ Programming Language, Special Edition; Addison-Wesley
Publishing Company, 2000.
11. Meyers Scott; More Effective C++; Addison-Wesley Publishing Company, 1996.
12. Irv Englander; The Architecture of Computer Hardware and Systems Software; 2nd Edition,
John Wiley & Sons, Inc., New York, 2000.
13. David J. Lilja; Measuring Computer Performance. A Practitioner’s Guide; Cambridge University
Press, 2000.
14. Arnold S. Berger; Hardware and Computer Organization—The Software Prospective; Elsevier,
Amsterdam, 2005.
❖ 626 ❖ Programming with C++
15. Internet Sources:–
a. cppreference.com
b. http://www. glenmccl.com/tmpl_cmp.html
c. http://microsoft.toddverbeek.com/lang.html
d. ttp://www.cplusplus.com/doc/tutorial
e. http://www.gnu.org/
f. http://gcc.gnu.org/
16. Peter Coad and Edward Yourdon; Object-Oriented Analysis; YourDoon Press, New Jersey,
1991.
17. Yashavant. P Kanetkar; Visual C++ Programming; BPB Publications, New Delhi, 1998.
18. Dimitris N. Chorafas; Fourth and Fifth Generation Programming Languages. Vol. II,
McGraw-Hill Book Company, 1986.
19. A. P. Ershov, Origins of Programming; (Translated by Robert H. Silverman); Springer-
Verlag, Berlin.
20. M. Ben-Ari; Principles of Concurrent Programming; Prentice-Hall.
21. Adele Goldberg; Small Talk; Addison-Wesley Publishing Company, California, USA, 1984.
22. Robert Lafore; Object-Oriented Programming in Turbo C++; Galgotia Publications Pvt. Ltd.,
1993.
23. Martin Richards, Colin Whitby – Streven; BCPL – The Language and its Compiler;
Cambridge University Press, London, 1980.
24. Michel Parent and Claude Laurgeau; Logic and Programming; Kogen Page, London.
25. Anthony J. Field and Peter G. Harrison; Functional Programming; Addison-Wesley
Publishing Company, California, USA, 1989.
26. Peter Abel; IBM PC Assembler Language and Programming; Prentice-Hall International,
Inc. 1987.
27. The C++ Standard; John Wiley Ltd., N.Y. 2003.
28. Danny Kalev; ANSI/ISO C++– Professional Programmer’s Handbook; Prentice-Hall of India
Pvt. Ltd., New Delhi-1999.
29. Paul E. Ceruzzi; A History of Modern Computing; The MIT Press, Cambridge, Massa-
chusetts, USA, 2002.
30. Mark Allen Weiss; Data Structures and Algorithm Analysis in C++; Pearson Education
(Singapore) Pvt. Ltd., Delhi-2005.

❍❍❍
 ASCII CHARACTER CODE SET

Decimal Binary Hexadecimal Character Description

0 0 0 NUL
1 1 01 SOH Start of heading
2 10 02 STX Start of text
3 11 03 ETX End of text
4 100 04 EOT End of transmission
5 101 05 ENQ Enquiry
6 110 06 ACK Acknowledge
7 111 07 BEL “\a”, bell, system beep
8 1000 08 BS “\b” , back space
9 1001 09 HT “\t” , horizontal tab
10 1010 0A LF “\n”, new line character
11 1011 0B VT “\v” , vertical tab
12 1100 0C FF “\f” , form feed
13 1101 0D CR “\r”, carriage return
14 1110 0E SO Shift out
15 1111 0F SI Shift in
16 10000 10 DLE Data link escape
17 10001 11 DC1 Device control 1
18 10010 12 DC2 Device control 2
19 10011 13 DC3 Device control 3
20 10100 14 DC4 Device control 4
21 10101 15 NAK Negative acknowledgement
22 10110 16 SYN Synchronous idle
23 10111 17 ETB End of transmission block
24 11000 18 CAN Cancel
25 11001 19 EM End of medium
26 11010 1A SUB Substitute
27 11011 1B ESC Escape
❖ 628 ❖ Programming with C++

Decimal Binary Hexadecimal Character Description

28 11100 1C FS File separator
29 11101 1D GS Group separator
30 11110 1E RS Record separator
31 11111 1F US Unit separator
32 100000 20 SPC Space

33 100001 21 ! Exclamation mark

34 100010 22 “ Double quote
35 100011 23 # Hash symbol
36 100100 24 $ Dollar sign
37 100101 25 % Percent sign
38 100110 26 & Ampersand
39 100111 27 , Single quote
40 101000 28 ( Left parentheses
41 101001 29 ) Right parentheses
42 101010 2A * Asterisk
43 101011 2B + Plus sign
44 101100 2C , comma
45 101101 2D - Dash
46 101110 2E . Dot, Decimal point
47 101111 2F / Slash
48 110000 30 0
49 110001 31 1
50 110010 32 2
51 110011 33 3
52 110100 34 4
53 110101 35 5 digits
54 110110 36 6
55 110111 37 7
56 111000 38 8
57 111001 39 9
58 111010 3A : Colon
59 111011 3B ; Semicolon
60 111100 3C < Less than
61 111101 3D = Equal to
62 111110 3E > Greater than
63 111111 3F ? Question mark
64 1000000 40 @ At the rate of symbol
 Appendix ❖ 629❖

Decimal Binary Hexadecimal Character Description

65 1000001 41 A Capital letters
66 1000010 42 B
67 1000011 43 C
68 1000100 44 D
69 1000101 45 E
70 1000110 46 F
71 1000111 47 G
72 1001000 48 H
73 1001001 49 I
74 1001010 4A J
75 1001011 4B K
76 1001100 4C L
77 1001101 4D M
78 1001110 4E N
79 1001111 4F O
80 1010000 50 P
81 1010001 51 Q
82 1010010 52 R
83 1010011 53 S
84 1010100 54 T
85 1010101 55 U
86 1010110 56 V
87 1010111 57 W
88 1011000 58 X
89 1011001 59 Y
90 1011010 5A Z

91 1011011 5B [ Left square bracket

92 1011100 5C \ Backslash
93 1011101 5D ] Right square bracket
94 1011110 5E ^ Caret Mark
95 1011111 5F _ Under score
96 1100000 60 ‘ Accent mark

97 1100001 61 a lower case letters

98 1100010 62 b
99 1100011 63 c
100 1100100 64 d
101 1100101 65 e
102 1100110 66 f
❖ 630 ❖ Programming with C++

Decimal Binary Hexadecimal Character Description

103 1100111 67 g
104 1101000 68 h
105 1101001 69 i
106 1101010 6A j
107 1101011 6B k
108 1101100 6C l
109 1101101 6D m
110 1101110 6E n
111 1101111 6F o
112 1110000 70 p
113 1110001 71 q Lower case letters
114 1110010 72 r
115 1110011 73 s
116 1110100 74 t
117 1110101 75 u
118 1110110 76 v
119 1110111 77 w
120 1111000 78 x
121 1111001 79 y
122 1111010 7A z
123 1111011 7B { Left curly bracket (brace)
124 1111100 7C | Pipe sign
125 1111101 7D } Right Curly bracket
126 1111110 7E ~ Tilde mark
127 1111111 7F DEL delete
 Appendix ❖ 631❖

C++ KEYWORDS

Keyword Description
and A synonym for the Boolean operator AND
and_eq Bitwise AND assignment. A synonym for &= operator
asm Specifies that the code be passed on to assembler directly.
auto Storage class that is used to define variables within a block{}, i.e. local variables
bitand Bitwise AND. A synonym for & operator.
bitor Bitwise OR. A synonym for | operator.
bool bool type specifier. Can have values 1 or 0 ( true or false )
break Used to terminate a loop. Also used in switch statement for the same purpose.
case Used in switch statement to specify a match.
catch Catch (T) specifies a catch block for type T. The block takes action when the
T type exception occurs.
char Type used to define character objects.
class Used for class declaration. It creates user defined type.
compl A synonym for bitwise NOT
const Specifies objects that do not change value for the lifetime of program.
const_cast Provides access to an object with constant or volatile attribute.
continue Used to transfer control to the beginning of loop.
default Used in switch statement to handle the statement not dealt by any case.
delete Releases the memory allocated by new operator.
do Indicates the start of do –while loop
double A fundamental data type used to define decimal point numbers with double
precision.
dynamic_cast Used to cast a pointer or reference type at run time.
else Specifies an alternative choice in if – else
enum Enumeration type
explicit Declares explicit constructor
export Makes a template definition accessible from another compiler unit
extern Specifies a storage class object with external link to the block
Contd...
❖ 632 ❖ Programming with C++

Key word Description

false One of the two values, i.e. (0) of bool type
float A fundamental type for specifying floating point numbers
for Specifies start of for loop
friend A class or function specified as friend has access to private , protected
and public members of the class.
goto Used to transfer control to a label statement.
if Specifies if expression for selective control
inline Specifies a function for inline substitution instead of function call.
int A fundamental type which specifies whole numbers.
long A data modifier for int and double
mutable Specifier for overriding constantness
namespace Specifies a scope for names
new Operator for allocating memory
not A synonym for ! operator, i.e. NOT
not_eq Not equal to. A synonym for !=.
operator Used for operator overloading
or OR operator. A synonym for ||.
Or_eq Bitwise OR assignment. A synonym for |= operator
private Specifies private members of a class
protected Specifies protected members of class
public Specifies class members which are accessible from outside the class
register An auto specifier for objects to be stored in register
reinterpret_cast Used to convert an int type to pointer type and vice versa
return Specifies end of function with a return value.
short A data type modifier for int numbers.
signed A data type modifier for numbers which can be positive or negative.
sizeof The sizeof() returns size of an object in number of bytes allocated for
object.
static Specifies objects that stay throughout the program
static_cast Used for casting. static_cast respects the constantness of object while
casting.
struct Declares a user type
switch Specifies switch statement for multiple choices.
template Specifies a generic type

Contd...
 Appendix ❖ 633❖

Key word Description

this A class pointer by compiler that points to class object.
throw Generates an exception
true Specifies a bool condition with value 1
try Indicates start of exception handling block.
typedef Declares a synonym for a type (fundamental type or user defined type)
typeid typeid ( ) returns the type of its argument
typename Used as an alternative to word class in template.
using Used as declaration or directive.
union It specifies a structure in which the members occupy the same memory
space.
unsigned A type modifier which specifies that all the bits are to be used for the
object, i.e. positive value.
virtual Specifies a member function which will be redefined in derived class.
void Signifies absence of type of function or absence of parameter list
volatile Declares an object that may be modified outside the control of program
(undetected by compiler)
wchar_t Wide character type allocated 2 byte memory
while Specifies start of while loop and end of do – while loop
xor Bitwise exclusive OR operator
xor_eq Bitwise exclusive OR assignment operator
❖ 634 ❖ Programming with C++

C++ OPERATORS
The operators are shown below in order of decreasing order of precedence. Those in the
same group have same precedence. In case of multiple occurrences in same code line
the associativity takes care of order of implementation.
Operator Arity Description Associativity
:: Unary Global scope Right to left
:: binary Class scope Left to right
. Binary Object component selector Left to right
() Not applicable Function call Left to right
() Not applicable Brackets – grouping operator Left to right
[] Binary Array index operator Left to right
-> Binary Access from a pointer Left to right
! Unary Logical NOT
~ Unary Bitwise compliment
++ Unary Increment
–– Unary Decrement
– Unary Unary minus
+ Unary Unary plus
* Unary Dereference
& Unary Address_of operator Right to left
Sizeof() Size in bytes
new Dynamic memory allocation
new[] Dynamic memory allocation
delete Dynamic memory deallocation
delete[] Dynamic memory deallocation
(type) Binary C-type casting to given type Right to left
conversion
–>* Binary Pointer to member via pointer Left to right
.* Binary pointer to member via dot selector
Contd...
 Appendix ❖ 635❖

Operator Arity Description Associativity

* Binary Multiplication
/ Binary Division Left to right
% Binary Modulus
+ Binary Addition Left to right
– Binary Subtraction
<< Binary Bitwise shift left Left to right
>> Binary Bitwise shift right
< Binary Less than
> Binary Greater than Left to right
<= Binary Less than equal to
>= Binary Greater than equal to
== Binary Equality Left to right
!= Binary Inequality
& Binary Bitwise AND Left to right
^ Binary Bitwise XOR Left to right
| Binary Bitwise OR Left to right
&& Binary Logical AND Left to right
|| Binary Logical OR Left to right
?: Binary Conditional operator Left to right
= Binary Assignment operator
+= Binary Add and assign
–= Binary Subtract and assign
*= Binary Multiply and assign
/= Binary Divide and assign Right to left
%= Binary Modulus and assign remainder
&= Binary Bitwise AND and assign
^= Binary Bitwise exclusive OR and assign
|= Binary Bitwise OR and assign
<<= Binary Bitwise left shift and assign
>>= Binary Bitwise right shift and assign
, Binary Evaluation operator Left to right
❖ 636 ❖ Programming with C++

COMMONLY USED HEADER FILES IN C++

S.No. Name Description
1 <algorithm> Contains algorithms- functions for manipulation of element of C++
Standard Library containers.
3 <bitset> Bitset classes
4 <cassert> C-style assert () macro.
5 <cctype> Testing character for certain properties.
6 <cerrno> Error handling – C-style
7 <cfloat> Floating point implementation classes
8 <cheks.h> For diagnostic.
9 <climits> Size limits for integral types of the system. It replaces <limits.h>
10 <clocale> Country and language specific
11 <cmath> Mathematical functions
12 <complex.h> Complex number classes
13 <constrea.h> Supports consol output
14 <csetjmp> Contains setjump() and longjump() which enable immediate jump
from deeply nested function call.
15 <csignal> C-style processing of signals
16 <cstdarg> Support for variable length function arguments
17 <cstddef> C-style library support.
18 <cstdio> Support C-style standard input/output. It replaces <stdio.h>
19 <cstdlib> Definitions for commonly used functions and types
20 <cstring> C-style string manipulation functions.
21 <ctime> Time and date manipulating functions.
22 <cwchar> Supports wide character
23 <deque> Container classes for double ended queues.

Contd...
 Appendix ❖ 637❖

24 <exception> Exception handling classes.

25 <fstream> Contains classes and functions for file input/output streams
26 <functional> Supports function objects.
27 <iomanip> Contains functions for manipulation of input /output streams.
28 <ios.h> Basic-stream input/output classes
29 <iosfwd> Support for forward input/output declarations.
30 <iostream> Supports standard input/output streams.
31 <istream> Supports standard input streams.
32 <iterator> Supports iterators used for traversing elements of a container.
33 <limits> Classes for numerical data type limits
34 <list> Container classes for doubly linked list of type T
35 <locale> Comprises classes and functions to support stream processing
for different languages.
36 <map> Associative container classes for maps and multimaps
37 <memory> Contains classes for memory allocation
38 <new> or <new.h> Supports allocation and de-allocation of memory
39 <numeric> Numerical manipulation algorithms
40 <ostream> Supports output stream
41 <queue> Container classes for queues and priority queues.
42 <ref.h> Some classes for strings. Some compilers may not have it.
43 <regexp.h> Supports regular expression search
44 <set> or <set.h> Associative container classes for set and multiset .
45 <sstream> Contains functions and classes for input/output of strings.
46 <stack> Adapter class stack
47 <stdexcept> Exception handling classes of C++ Standard Library
48 <streambuf> Stream buffer classes
49 <string> or Deals with C++ string classes.
<string.h>
50 <strstrea.h> Stream classes dealing with byte arrays.
51 <typeinfo> or Run time type identification classes.
<typeinfo.h>
52 <utility> General utility templates used by many C++ std library classes.
53 <valarray> For numerical manipulation of vectors
54 <vector> Vector container classes
❖ 638 ❖ Programming with C++

GETTING STARTED
If you are using Microsoft Visual Studio 6.0, below is the step by step illustration of how
to run a C++ program. In the first case we run a simple program without a class and in the
second case we create a project, include a class and implement a program.
Load the Microsoft Visual Studio 6.0 on your computer. Restart your computer and click
on Programs menu. Out of the list of programs select Microsoft Visual C++ 6.0. You will
get the screen as shown below.

Fig.1

It has three compartments. A view pan on the left. A document window on the right
and a message window at bottom. The message window is used for displaying error messages
or warnings when the program is compiled. The document widow is used for typing the
program. When a class program is run the left view gives option for viewing the class or file
view by hitting on the appropriate button provided at its bottom (see Fig.9). For the first
case, described below, we will not need it.
 Appendix ❖ 639❖
CASE 1 – RUNNING A SIMPLE PROGRAM
Now open the file menu and select (click with mouse) on New. You will see the screen as
demonstrated in Fig. 6 below. In the new window click on files. You would get the new
window as shown in Fig. 2. Select (click) on C++ source file. It would get shaded as shown
in Fig. 2. In the ‘File name’ window, type the name of file. In the present case it is typed
Myprogram with extension .cpp. So complete name typed is Myprogram.cpp. In the location
window in the present case, it is filled by default. Click/hit on OK. A dialog box will appear
for creating a project. Click on ‘yes’.

Fig.2
You will get a blank right document window on which you type the statements of your
program as illustrated in Fig. 3. The following program has been typed. You may close the
window pane by hitting (Clicking) on x on top right corner of view pane.
#include<iostream>
using namespace std;
int main()
{
cout<<“Welcome to C++!”<<endl;
return 0;
}

After typing the program, open the ‘Build’ menu and select (click on) ‘Compile
Myprogram.cpp’.
The program gets compiled and if there are any errors in the program, these would be
shown in the bottom window. You can correct them and again compile the program.
❖ 640 ❖ Programming with C++

Fig. 3

Fig. 4
When it becomes error free you would get the screen as illustrated in Fig.4, in which the
bottom window shows,
0 error(s), 0 warning(s).

The next step is to execute the program. For this open ‘Build’ menu and click on ‘Execute
Myprogram.exe. The result of this is shown in the dark window in Fig.5
 Appendix ❖ 641❖

Fig. 5

The program with the output is shown below.

#include<iostream>
using namespace std;
int main()
{ cout<<“Wecome to C++!” << endl;
return 0;
}
The output is given below.
Wecome to C++!
Press any key to continue

CASE 2 – MAKING A PROGRAM WITH A CLASS

As mentioned in Case 1, start your computer and click on Programs menu. Out of the list
of programs select Microsoft Visual C++6.0. You will get the screen as shown in Fig.1. Open
‘File’ menu and click on New. You get a screen as illustrated in Fig.6. In the ‘Project name’
window type the name of project. In the present case the name is ‘Training’. In the location
window you may type the place and path where you want to store the project. In the present
case it is a default site.
❖ 642 ❖ Programming with C++

Fig. 6
Out of the list of options select ‘Win32 Console Application’ and the ‘Create new work-
space’ is selected by default, if not, select it. After filling these click on OK. You get a screen
as shown in Fig.7. Just click on ‘Finish’. You may get a message page which tells you that
you have created an empty project. Click on OK. Now you get a screen similar to the Fig.1
but with a difference that in the left view pane the name ‘Training’ appears. Our next step is
to add a class to the project.
Again open File menu and click on New. You get a screen as shown in Fig.8. In this you
will find that the name of project is already filled by default and ‘Add to project’ is also
selected if not you can select it.

Fig. 7
 Appendix ❖ 643❖

Fig. 8

Now we have to add a class to the project. A class needs a header file. So select ‘C/C++
Header File’ from the list of options. It is shown shaded in Fig.8. In the ‘File name’ window
(Fig. 8) fill the name of class. In the present case it is typed as ‘Rect.h’ . The ‘Location’ window
is filled by default. After filling the name click on OK. You get the screen in which the right
window is blank. In this window type the statements of the class as illustrated in Fig.9.

Fig. 9
❖ 644 ❖ Programming with C++

The class entered is as below.

#ifndef Rect_h
#define Rect_h
class Myrect
{ pubic:
void Setvalues(int A, int B)
{ x = A; y = B;}
int Area ()
{return x*y;}
private:
int x, y;
};
#endif

Now save this. After typing the class statements the message window would be blank.
The messages given in the third window at bottom of Fig.9 is in fact the result of next
operation. So, there should be no confusion.
After typing and saving the next step in implementation of class. For this again open file
menu and click on New. The screen is illustrated in Fig.10. Now select ‘C++ Source File’
from the list.

Fig. 10
You will find that ‘Add to project’ is selected by default, the name of project “Training”
also appears by default. You have to give a name to this file. All the files must have distinct
names. So for this file we type the name ‘Rectangle.cpp’ as illustrated in the figure. The window
below Location is also filled by default. After writing the name click on OK. Again you will
get screen with three segments and the blank right window. Here you type the main program
for implementing the class. This is illustrated in Fig.11 which also illustrates the output which
is the result of subsequent operations. The program typed in this window is given further.
 Appendix ❖ 645❖
#include<iostream>
using namespace std;
#include “Rect.h”
int main()
{MyRect Rect1, Rect2;
Rect1.Setvalues(15,10);
Rect2.Setvalues(20,5);
cout<<“Area of Rect1 = ”<<Rect1.Area()<<endl;
cout<<“Area of Rect2 = ”<<Rect2.Area()<<endl;
return 0;
}
After typing the program, open the Build menu and click on Compile option. If there
are errors these would be shown in the message window at the bottom, otherwise the message
would be no error(s), no warnings. In the program Rect1 and Rect2 are two objects of
MyRect. So they are declared as
MyRect Rect1, Rect2;

Here Rect1, Rect2 are of type MyRect. The function Setvalues() is called by Rect1 and
by Rect2 by the expressions
Rect1.Setvalues(15,10);
Rect2.Setvalues(20,5);
The Rect1 has dimensions 15 and 10 and Rect2 has dimensions 20 and 5. Next the
Area () function for the two objects is called. Again open the Build menu and now click
for execution of the program by clicking on Execute Rectangle.exe . You would get the result
as shown on the dark window in Fig. 11. The output shows the areas as 150 and 100.

Fig. 11
❖ 646 ❖ Programming with C++

The execution is now over and program may be closed. Later if you want to go back to
this project select Visual C++6.0 and open the File menu. And click on ‘Open’. You would
get the screen as given in Fig.12. You may open the class Rect.h or the file Rectangle.cpp as
desired.

Fig. 12

❍❍❍
A by value 160,
by pointers 224
abort (a program) 431 Arithmetic and logic unit 1
Absolute value 159 Arithmetic assignment operator 77
Abstract class 377 Arithmetic operators 73
Access specifiers for class 264 precedence 83-84, 634
Access private class data 302,269 Arrays 167
accumulate() , STL 592, 593 Accessing elements 169
acos(), <cmath> 159 address 179
Address of class object 305 binary search 178
Address-of operator (&) 197, 201 declaration 167
adjacent_difference (), STL 594 finding max / min values 186
adjacent_find() , STL 589, 594 initialization 167-168
algorithms (STL) 588 input / output 170
mutating 591 memory 179
non mutating 590 multidimensional 168
numeric 593 name as const. pointer 205
on heap 594 of pointers 214
on sets 593 of strings 244
permutation 593 searching a value 177
relational 590 sorting 183
search and sort 592 sort () function 592
<algorithm> header file 588 Appendix A 627
ANSI 4 Appendix B 631
AND 90 Appendix C 634
bitwise operator 90 Appendix D 636
logical operator 90 Appendix E 638
keyword 59 asm (keyword ) 59
any (), <bitset> 584 Assembler 3
Arguments Assembly language 3
by reference 160, assert () 506
❖ 648 ❖ Programming with C++

Assignment operators 77 -81 Binary files 484

ASCII ( American code for information binary_search 592, 595
interchange) 53, also see Appendix A Binder function () 529
assign() (STL) 535, 553 Bit and bytes 20
associative containers 517 Bit fields 469
associative container functions 559 bitand ( keyword) 59
associativity 77 , also see Appendix C bitor ( keyword) 59
of operators see Appendix C bitsets 577
asterisk (*) construction of 579
multiplication operator 80 header file 578
dereference operator 197 functions 584
pointer 197 any () 584
at() member function 535 count() 584
of deque 553 flip () 584
of vector class 535 none ()584
atof () function 241 reset () 584
atoi ()function 241 set () 584
atol () function 241 size ()584
auto (keyword) 59, 64 test () 584
auto_ptr class 450 to_string () 584
automatic storage class 66 to_ulong () 584
bitwise operators 581
B AND operator 581
Complement 581
B- programming language 3
Inclusive OR 581
back () , STL , 535 553
Exclusive OR 581
bad() 489
Left shift 581
bad_alloc 447
Right shift 581
badbit 489
XOR assign 581
bad_cast 447
bool data type 54
bad_exception 447
Boolean logical operators 80, 90
bad_typeid, exception class, 447
Borland 8
base class 16, 339
break statement 126
base 18
Byte 20
BASIC Language 4
basic_fstream 472
basic_ifstream 472
C
basic_ios class 454 C language 4
basic_iostream class 455 C-90 4
basic_istream class 455 C-99 4
basic_ofstream class 472 C++ 98 5
basic_ostream calss 455 C++
BCPL programming language 4 Advantages of C++ over C 6
begin () STL 535, 553 Compilers 8
Better C 5 History 4
Bidirectional iterator 513 Language 5
Binary number system 17 Preprocessor 501
 Subject Index ❖ 649❖

Range of Applications 7 containers 510, 519

Standard Library 8 adapters 519
<string> header file 417 associative 518
Cache memory 23 deque 515
Calling a function 148 list 516
capacity (), STL, 535 vector 513
case (keyword) 59 Containment 361
casting (type) 68 constants 63
catch() 432 string constant 63
all exceptions (…) 438 control unit 1
keyword 59 copy (), STL Algorithm, 591
CD ROM 1 copy_backward () 591
ceil () 159 count () (bitset class) 584
cerr () 455 count (), STL , Algorithm 590
char data type 54 count_if (),STL , Algorithm 590
character 56 CPU 1
cin 455 C- strings 231
cin.get() 235
cin.getline() 45, 238 D
cin.ignore () 240
Data abstraction 13
cin.peek() 240
Data hiding 15
cin,putback () 239-240
Data member 263
class
Data types 34, 54
abstract 377
dec 464, 467
access specifiers 264
Decrement operator see operators
constructor and destructor 277
default keyword 59
copy constructor 278
delete keyword 59
default 278
delete [] 216
parametric 278
deque 515
derived class 262,
dereference operator 197, 201
friend function 291
destructor 277
friend class 294
digit constants 63
local 283
do - - - while 123
matrix 312
double 34, 54
nested 320
Doubly linked list 314-319, 516
pointer 296
Dynamic binding 369
static data member 307
dynamic_cast <>() 388
static function member 308
Dynamic memory allocation 216
clear () (STL) 535 , 553
clear() , file operation 489
clog 455 E
close() 473 Early binding 369
Comments 31 else (keyword) 59
Composite assignment operators 8 end() 553
Computer 1 endl 28, 464
const 59, 63 ends 464
const_cast <>() 70, 393 endless loops 126, 129
❖ 650 ❖ Programming with C++

enum 54 find(), STL, 590 , 602

eof () 489 find_end (), STL, 590 , 603
eofbit 489 find_first_of() ,STL, 590 , 604
empty() , STL , 535, 553 find_if (), STL, 590, 604
equal () STL Algorithm, 590 fixed () 467
equal_range() , STL Algorithm 590, 610 flags 466
equal_to () 590 flip flop 20
erase() , STL 535, 553 flip(),<bitset> 584
extern 64 float 34, 54
vector (container class) 533 floor() 159
Escape sequence 30-31 floppy disc 1
Evaluation order 77-78, 84, 95 flush 464
Exception handling 431 for (keyword) 59
exception classes 446 for loop 127
nested loop 131
exception specification 441
for_each () , STL Algorithm
exception handling function 440
Formatting flags 466
exit() 126
Formatting I/O 464, 466
explicit ( keyword) 59
Formatting output 39
extern 64
Fractional numbers 55
extraction operator >> 33, 453 free() 216
friend
F keyword 59
fabs() 159 class 294
Factorial of a number 125, function 291
fail() 489 inheritance 367
failbit 489 Fortran 4
false 57, 90 front () , STL, 535, 553
Function
formatting output 464
C++ Standard Library
File
Arguments 144-147
access 470,
Body 146
binary 484
Calling 148
disc file 470
Declaration 147
eof () 489
definition 145-148
input / output file streams 473 inline 154
is_open() 478 pointer to 215
opening mode 480 object 526
operations 474 overloading 153
pointer 481 passing values by reference 160,
random access 488 template 396
read() 485 type 144-145
stream classes 472 value 148
sequential 470 void 153
write() 485 Function prototype 147
fill() member function 466 Function return type 144-149
fill() STL 591, 601 return value 148
fill_n () , STL, 591, 601 Fundamental data type 53
 Subject Index ❖ 651❖

G Interrupt 1
istream 453- 456
get() 458, 478 invalid argument 447
getline () function for cin 461 ios_base
getline () function of string header file ios:: adjustfield
good() 489 basefield 467
goodbit 489 floatfield 467
goto statement 136 app , (file operation) 480
global variables 66 ate , (file operation) 480
beg 482, 525
H binary , (file operation) 480
Hard disc 1 cur 482
Header file 28 dec 464, 467
Hexadecimal number system 18 end 482
High level language 3 formating flags 466
hex 464, 467 fixed 467
hex 464, 467
I in 480
internal 467
If left 467
Keyword 59 nocreate (file operation) 480
(expression) 100 noreplace (file operation) 480
if - else 102 oct 464
ifstream 472-473 out ( file operation) 480
ignore () 461 right 467
includes(), STL, algorithm 593 scientific 467
increment operator see operators showbase 469
Index 167 showpoint 469
Identifier for a variable 58 showpos 469
Indirection operator 197 skipws 469
Information hiding 264 trunc , (file operation) 480
Inheritance unitbuf 469
access declaration 339 uppercase 469
access specifiers 342 350-351 ios 454, 467
\containment 361 iostream 453-455
constructor and destructor in 355 istream iterator 454
default 342 is_open() 478
private 342, 348 isalnum() 246
protected 342, 346 isalpha() 246
public 342 iscntrl() 246
declaration 339 isdigit() 246
multiple 351 islower() 246
multilevel 352 ispunct() 246
operator overloading 366 isspace() 246
inner_product (), STL,593, 607 isupper() 246
inplace_merge() STL, 592, 611 iter_swap(), STL, algoritm 591, 608
insert () , STL, 535, 553 iterator
❖ 652 ❖ Programming with C++

input 512-513 malloc() 216

output 512-513 manipulators
bidirectional 512-513 endl 464
forward 512-513 flush 464
random 512-513 hex 464
non-parameterised 464
J oct 464
parameterised 465
Java 24
resetiosflags() 465
setiosflags () 465
K setpresision() 465
Keywords in C++ 59 setw() 465
map 559, 567
L Matrix class 312
max_size (),STL, 535, 553
Left justification 41, 467 memcmp() 256
length_error 447 memchr() 256
Lexicographical_compare() STL, 590 memcpy() 256
Linked list 314 Memories in computer 23
List 510, 532, 546 memmove() 256
logic_error 446 Memory Functions of C-strings 256
logical false, true 57, 90 memset() 256
logical AND 80, 90 merge() STL, 546
logical NOT 80, 90 min_element (), STL, 590, 612
logical OR 80, 90 mismatch(), STL, algorithm, 590, 612
long 54 , 59 Modular programming 11
loop modulus 82
do- while 123 Monolithic programming 10
endless 126, Multimap 511, 574
for 127 multiple inheritance see inheritance
nested 119, multiset 511, 566
while 117 mutable 64
lower_bound (), STL, 592, 609
N
M
Namespaces 36, 493
Macros 507 alias 496
predefined preprocessor application 37
_cplusplus 507 extension 498
_DATE_ 507 nesting 495
_FILE_ 507 std 500
_LINE_ 507 using 495
_STDC_ 507 Nested
_TIME_ 507 class 320
Machine language 3 for loop 131
main() 29 while loop 119
make_heap (), STL, 594 new (keyword) 59
 Subject Index ❖ 653❖

new [ ] 216 P
next_permutation(), STL, 592
nocreate, iso(file operation) 480 partial_sort() 592
noreplace, ios (file operation)480 partial_sort_copy() 592
none() (bitset class) 584 partial_sum() 593
Null 231 partition() 592
NULL 198 peek() 240
Number systems 17 Pointer
Arithmetic 226
O array of 214
casting 391
Object const 205
definition 263 declaration of 197
Object oriented programming 12, 262 decrement 226
oct 464 function argument 215
Octal number system 18 increment 226
ofstream 472 NULL 198
open () 473, 480 Passing arguments through 224
operators , also see appendix C
smart see auto_ptr class
arithmetic 82
this 305, 328
arity 77
to array 205, 208, 210
associativity 77
to class 296
bitwise 80
to class objects 299
Boolean 80, 90
to data member 301
Precedence 77
to function 212
Operating system 4
to member function 300
Operator
to pointer 203
Arithmetic 80, 82
void 227
array subscript operator 167, 335, 634
Polymorphism 16, 369
assignment 80
pop_back () 535, 553
binary 77
pop_front () 546, 553
bitwise 80
bool 54 pop_heap () 594
complement 80 Portability 3
composite operators 80 precision()42, 465
decrement 80 Precedence 95, 634
increment 80 Predicates 526
delete 216 predefined 528
extraction operator >> preprocessor directives
dereference 197 # and ## 501
insertion operators << 28, 453 conditional directives
new 216 #define 501
overloading 322 -338 #elif 504
ternary 77 #else 504
ostream 453-455 #endif 504
out_of_range 447 #error 501
output_iterator 513 # if 504
overflow_error 447 # ifdef 504
❖ 654 ❖ Programming with C++

#ifndef 504 replace_copy () 590

# include 501 reserve () 535,
#line 501 resetiosflags() 465
# pragma 501 reset() ,(bitset class) 584
# undef 54 resize () 535, 553
prev_permutation () 593, 616 return 147-148
priority queue 525 return statement 148
private reverse(), algorithm, 546,
data member 267, 269 reverse_copy() 590
function member 281 right justification 41, 467
Procedural programming 10 rotate() 590, 619
Programming techniques 10 rotate_copy() 590
protected runtime_error 446
data member 267, 269 RTTI ( run-time type information) 385
function member 265, 343
public S
data member function 265
scientific notation 40, 467
function member 265
scope of variables 66-67
push_back() 535, 553
search() 590
push_front() 546, 553
search_n () 590
push_heap () 594
seek_dir 482
put() 458, 478
seekg() 482
putback() 240
seekp() 482
set () (bitset class) 584
Q set_intersection() 592, 620
queue 522 set_difference() 592, 620
queue priority 525 set_symmetric_difference()592, 620
set_terminate() 448
R set_unexpected() 449
set_union() 592, 620
RAM 1
setbase() 465
rand() 132
setf() 465
random_shuffle() 590, 616
setfill() 465
random file access 488
setiosflags() 465
rdstate() 489
setprecision() 465
rbegin () 535, 553
setw() 465
read() 463, 485
shift operators
References (for variable) 218,
<< left shift 581
References (Bibliography) 625
>> right shift 581
register 64
Singly linked list 314
reinterpret_cast<>() 391
short 60
Relational operators 55
showbase 469
remove(), algorithm , 546,
showpoint469
remove_copy() 590
showpos 469
remove_if () 546,
signed numbers 22
rend() 535 , 553
Simula 4
replace(),algorithm , 590, 618
 Subject Index ❖ 655❖

size() , STL, 535, 553, 584 strstr() 248

size_t 187 strtok() 248
sizeof () 61 streams
skipws 469 cerr 455
Smalltalk 4 cin 455
smart pointers see auto_ptr class clog 455
sort () 546 console 455
sort_heap() 594 cout 455
splice () 546 predefined 455
sqrt () 119 wcerr 455
srand() 132 wcin 455
stable_partition() 592 wclog 455
stable_sort() 592 wcout 455
stack 519 status flags 489
standard input/output 454 string
Standard manipulators 464 C-string 231
Standard Template Library C++ string 417
(STL) 509 struct 284
Standard functions swap() 535, 553
For string characters 246 swap_ranges()535, 622
For conversion of strings 241 switch 109-113
For string elements 239
Manipulation of C-strings 248 T
For C++ string 419 -421
tellp() 482
static
tellg() 482
data member 307
template
function member 308
class 407
variable 64
class for complex variables 415
static_cast<>() 70, 389
friend function 410
std ( namespace) 36, 493
function 396
strod() 242 -3
multipletype arguments 404
strtol() 242-3
overloading functions 406
strtoul () 242-3
single type arguments 396
storage class specifiers
specialisation 396
auto 64
terminate () 448
extern 64
test() bitset class 584
mutable 64
this pointer 305
static 64
throw (exception) 432
static_cast <>()
time () 134
strcat () 248
tolower() 246
strncat () 248
to_string() bitset class 584
strchr () 248
to_ulong () , bitset class 584
strcmp() 248
toupper() 246
strcpy() 248
transform() 591
strlen () 248
true 57, 90
strncpy() 248
try 432
strspn() 248
❖ 656 ❖ Programming with C++

type fundamental 54 destructor 379

type casting 70 pure 377
type_info (RTTI) 385 void 29, 34
typedef 71, 387 pointer 227
typeid 72, 387 volatile 63
typename 387
W
U wchar_t 54, 59
underflow_error 447 while loop 117
unexpected () 449 whitespace 32
unibuf 469 width() 41,
union ( keyword ) 59 wiostream 455
unique() 546, 591, 622 wistream 455
unique_copy () 691 wostream 455
unsetf() 466 Working with files 470
unsigned 55 wstreambuf 455
upper_bound (), STL, 591, 623 write () 453, 485
uppercase() 469 ws 464
User interactive programs 38 X
XOR operator 80, 580 , 634
V XOR_eq ( key word) 59
valarray 417
vector class 512, 531-544
Z
virtual Zeros and ones 1
base class 383 Zeroth element 167
functions 369, 373