Sanet - ST 1683923537python PDF

PYTHON
BASICS
LICENSE, DISCLAIMER OF LIABILITY, AND LIMITED WARRANTY
By purchasing or using this book (the “Work”), you agree that this license grants permission to use the
contents contained herein, but does not give you the right of ownership to any of the textual content in the
book or ownership to any of the information or products contained in it. This license does not permit
uploading of the Work onto the Internet or on a network (of any kind) without the written consent of the
Publisher. Duplication or dissemination of any text, code, simulations, images, etc., contained herein is
limited to and subject to licensing terms for the respective products, and permission must be obtained from
the Publisher or the owner of the content, etc., in order to reproduce or network any portion of the textual
material (in any media) that is contained in the Work.
MERCURY LEARNING AND INFORMATION (“MLI” or “the Publisher”) and anyone involved in the
creation, writing, production, accompanying algorithms, code, or computer programs (“the software”), and
any accompanying Web site or software of the Work, cannot and do not warrant the performance or results
that might be obtained by using the contents of the Work. The author, developers, and the Publisher have
used their best efforts to insure the accuracy and functionality of the textual material and/or programs
contained in this package; we, however, make no warranty of any kind, express or implied, regarding the
performance of these contents or programs. The Work is sold “as is” without warranty (except for defective
materials used in manufacturing the book or due to faulty workmanship).
The author, developers, and the publisher of any accompanying content, and anyone involved in the
composition, production, and manufacturing of this work will not be liable for damages of any kind arising
out of the use of (or the inability to use) the algorithms, source code, computer programs, or textual material
contained in this publication. This includes, but is not limited to, loss of revenue or profit, or other
incidental, physical, or consequential damages arising out of the use of this Work.
The sole remedy in the event of a claim of any kind is expressly limited to replacement of the book and only
at the discretion of the Publisher. The use of “implied warranty” and certain “exclusions” vary from state to
state, and might not apply to the purchaser of this product.
PYTHON
BASICS
H. Bhasin
MERCURY LEARNING AND INFORMATION

Dulles, Virginia
Boston, Massachusetts
New Delhi
Copyright ©2019 by MERCURY LEARNING AND INFORMATION LLC. All rights reserved.
ISBN: 978-1-683923-53-4. Reprinted and revised with permission.
Original Title and Copyright: Python for Beginners.

Copyright ©2019 by New Age International (P) Ltd. Publishers. All rights reserved.
ISBN: 978-93-86649-49-2
This publication, portions of it, or any accompanying software may not be reproduced in any way, stored in
a retrieval system of any type, or transmitted by any means, media, electronic display or mechanical
display, including, but not limited to, photocopy, recording, Internet postings, or scanning, without prior
permission in writing from the publisher.
Publisher: David Pallai

MERCURY LEARNING AND INFORMATION
22841 Quicksilver Drive
Dulles, VA 20166
[email protected]
www.merclearning.com
1-800-232-0223
H. Bhasin. Python Basics.

ISBN: 978-1-683923-53-4
The publisher recognizes and respects all marks used by companies, manufacturers, and developers as a
means to distinguish their products. All brand names and product names mentioned in this book are
trademarks or service marks of their respective companies. Any omission or misuse (of any kind) of service
marks or trademarks, etc. is not an attempt to infringe on the property of others.
Library of Congress Control Number: 2018962670

181920321 Printed on acid-free paper in the United States of America.
Our titles are available for adoption, license, or bulk purchase by institutions, corporations, etc. For
additional information, please contact the Customer Service Dept. at 800-232-0223(toll free).
All of our titles are available in digital format at authorcloudware.com and other digital vendors. The sole
obligation of MERCURY LEARNING AND INFORMATION to the purchaser is to replace the book,
based on defective materials or faulty workmanship, but not based on the operation or functionality of the
product.
To
My Mother
CONTENTS
Chapter 1: Introduction to Python

1.1 Introduction
1.2 Features of Python
1.2.1 Easy
1.2.2 Type and Run
1.2.3 Syntax
1.2.4 Mixing
1.2.5 Dynamic Typing
1.2.6 Built in Object Types
1.2.7 Numerous Libraries and Tools
1.2.8 Portable
1.2.9 Free
1.3 The Paradigms
1.3.1 Procedural
1.3.2 Object-Oriented
1.3.3 Functional
1.4 Chronology and Uses
1.4.1 Chronology
1.4.2 Uses
1.5 Installation of Anaconda
1.6 Conclusion
Chapter 2: Python Objects
2.1 Introduction
2.2 Basic Data Types Revisited
2.2.1 Fractions
2.3 Strings
2.4 Lists and Tuples
2.4.1 List
2.4.2 Tuples
2.4.3 Features of Tuples
2.5 Conclusion
Chapter 3: Conditional Statements
3.1 Introduction
3.2 if, if-else, and if-elif-else constructs
3.3 The if-elif-else Ladder
3.4 Logical Operators
3.5 The Ternary Operator
3.6 The get Construct
3.7 Examples
3.8 Conclusion
Chapter 4: Looping
4.1 Introduction
4.2 While
4.3 Patterns
4.4 Nesting and Applications of Loops in Lists
4.5 Conclusion
Chapter 5: Functions
5.1 Introduction
5.2 Features of a Function
5.2.1 Modular Programming
5.2.2 Reusability of Code
5.2.3 Manageability
5.3 Basic Terminology
5.3.1 Name of a Function
5.3.2 Arguments
5.3.3 Return Value
5.4 Definition and Invocation
5.4.1 Working
5.5 Types of Function
5.5.1 Advantage of Arguments
5.6 Implementing Search
5.7 Scope
5.8 Recursion
5.8.1 Rabbit Problem
5.8.2 Disadvantages of Using Recursion
5.9 Conclusion
Chapter 6: Iterations, Generators, and Comprehensions
6.1 Introduction
6.2 The Power of “For”
6.3 Iterators
6.4 Defining an Iterable Object
6.5 Generators
6.6 Comprehensions
6.7 Conclusion
Chapter 7: File Handling
7.1 Introduction
7.2 The File Handling Mechanism
7.3 The Open Function and File Access Modes
7.4 Python Functions for File Handling
7.4.1 The Essential Ones
7.4.2 The OS Methods
7.4.3 Miscellaneous Functions and File Attributes
7.5 Command Line Arguments
7.6 Implementation and Illustrations
7.7 Conclusion
Chapter 8: Strings
8.1 Introduction
8.2 The Use of “For” and “While”
8.3 String Operators
8.3.1 The Concatenation Operator (+)
8.3.2 The Replication Operator
8.3.3 The Membership Operator
8.4 Functions for String Handling
8.4.1 len()
8.4.2 Capitalize()
8.4.3 find()
8.4.4 count
8.4.5 Endswith()
8.4.6 Encode
8.4.7 Decode
8.4.8 Miscellaneous Functions
8.5 Conclusion
Chapter 9: Introduction to Object Oriented Paradigm
9.1 Introduction
9.2 Creating New Types
9.3 Attributes and Functions
9.3.1 Attributes
9.3.2 Functions
9.4 Elements of Object-Oriented Programming
9.4.1 Class
9.4.2 Object
9.4.3 Encapsulation
9.4.4 Data Hiding
9.4.5 Inheritance
9.4.6 Polymorphism
9.4.7 Reusability
9.5 Conclusion
Chapter 10: Classes and Objects
10.1 Introduction to Classes
10.2 Defining a Class
10.3 Creating an Object
10.4 Scope of Data Members
10.5 Nesting
10.6 Constructor
10.7 Constructor Overloading
10.8 Destructors
10.9 Conclusion
Chapter 11: Inheritance
11.1 Introduction to Inheritance and Composition
11.1.1 Inheritance and Methods
11.1.2 Composition
11.2 Inheritance: Importance and Types
11.2.1 Need for Inheritance
11.2.2 Types of Inheritance
11.3 Methods
11.3.1 Bound Methods
11.3.2 Unbound Method
11.3.3 Methods are Callable Objects
11.3.4 The Importance and Usage of Super
11.3.5 Calling the Base Class Function Using Super
11.4 Search in Inheritance Tree
11.5 Class Interface and Abstract Classes
11.6 Conclusion
Chapter 12: Operator Overloading
12.1 Introduction
12.2 _init_ Revisited
12.2.1 Overloading _init_ (sort of)
12.3 Methods for Overloading Binary Operators
12.4 Overloading Binary Operators: The Fraction Example
12.5 Overloading the += Operator
12.6 Overloading the > and < Operators
12.7 Overloading the _boolEan_ Operators: Precedence of
_bool_over _len_
12.8 Destructors
12.9 Conclusion
Chapter 13: Exception Handling
13.1 Introduction
13.2 Importance and Mechanism
13.2.1 An Example of Try/Catch
13.2.2 Manually Raising Exceptions
13.3 Built-In Exceptions in Python
13.4 The Process
13.4.1 Exception Handling: Try/Except
13.4.2 Raising Exceptions
13.5 Crafting User Defined Exceptions
13.6 An Example of Exception Handling
13.7 Conclusion
Chapter 14: Introduction to Data Structures
14.1 Introduction
14.2 Abstract Data Type
14.3 Algorithms
14.4 Arrays
14.5 Iterative and Recursive Algorithms
14.5.1 Iterative Algorithms
14.5.2 Recursive Algorithms
14.6 Conclusion
Chapter 15: Stacks and Queues
15.1 Introduction
15.2 Stack
15.3 Dynamic Implementation of Stacks
15.4 Dynamic Implementation: Another Way
15.5 Applications of Stacks
15.5.1 Reversing a String
15.5.2 Infix, Prefix, and Postfix Expressions
15.6 Queue
15.7 Conclusion
Chapter 16: Linked Lists
16.1 Introduction
16.2 Operations
16.3 Implementing Stack Using a Linked List
16.4 Queue Using a Linked List
16.5 Conclusion
Chapter 17: Binary Search Trees
17.1 Introduction
17.2 Definition and Terminology
17.2.1 Graphs: Definition and Representation
17.2.2 Trees: Definition, Classification, and Representation
17.2.3 Representation of a Binary Tree
17.2.4 Tree Traversal: In-order, Pre-order, and Post-order
17.3 Binary Search Tree
17.3.1 Creation and Insertion
17.3.2 Traversal
17.3.3 Maximum and Minimum Elements
17.4 Conclusion
Chapter 18: Introduction to NUMPY
18.1 Introduction
18.2 Introduction to NumPy and Creation of a Basic Array
18.3 Functions for Generating Sequences
18.3.1 arange()
18.3.2 linspace()
18.3.3 logspace()
18.4 Aggregate Functions
18.5 Broadcasting
18.6 Structured Arrays
18.7 Conclusion
Chapter 19: Introduction to MATPLOTLIB
19.1 Introduction
19.2 The Plot Function
19.3 Subplots
19.4 3 Dimensional Plotting
19.5 Conclusion
Chapter 20: Introduction to Image Processing
20.1 Introduction
20.2 Opening, Reading, and Writing an Image
20.2.1 Opening an Image
20.2.2 Reading
20.2.3 Writing an Image to a File
20.2.4 Displaying an Image
20.3 The Contour Function
20.4 Clipping
20.5 Statistical Information of an Image
20.6 Basic Transformation
20.6.1 Translation
20.6.2 Rotation
20.6.3 Scaling
20.7 Conclusion
Appendix A: Multithreading in Python
Appendix B: Regular Expressions
Appendix C: Exercises for Practice: Programming Questions
Appendix D: Problems for Practice: Multiple Choice Questions
Appendix E: Answer to the Multiple Choice Questions
Bibliography
Index
CHAPTER
1
INTRODUCTION TO PYTHON
After reading this chapter, the reader will be able to
• Understand the chronology of Python
• Appreciate the importance and features of Python
• Discover the areas in which Python can be used
• Install Anaconda
1.1 INTRODUCTION
Art is an expression of human creative skill, hence programming is an art. The
choice of programming language is, therefore, important. This book introduces
Python, which will help you to become a great artist. A. J. Perlis, who was a
professor at the Purdue University, and who was the recipient of the first Turing
award, stated
“A language that doesn’t affect the way you think about programming is
not worth knowing.”
Python is worth knowing. Learning Python will not only motivate you to do
highly complex tasks in the simplest manners but will also demolish the myths
of conventional programming paradigms. It is a language which will change the
way you program and hence look at a problem.
Python is a strong, procedural, object-oriented, functional language crafted in the

late 1980s by Guido Van Rossum. The language is named after Monty Python, a
comedy group. The language is currently being used in diverse application
domains. These include software development, web development, Desktop GUI
development, education, and scientific applications. So, it spans almost all the
facets of development. Its popularity is primarily owing to its simplicity and
robustness, though there are many other factors too which are discussed in the
chapters that follow.
There are many third party modules for accomplishing the above tasks. For
example Django, an immensely popular Web framework dedicated to clean and
fast development, is developed on Python. This, along with the support for
HTML, E-mails, FTP, etc., makes it a good choice for web development.
Third party libraries are also available for software development. One of the
most common examples is Scions, which is used for build controls. When joined
with the inbuilt features and support, Python also works miracles for GUI
development and for developing mobile applications, e.g., Kivy is used for
developing multi-touch applications.
Python also finds its applications in scientific analysis. SciPy is used for
Engineering and Mathematics, and IPython is used for parallel computing. Those
of you working in statistics and machine learning would find some of these
libraries extremely useful and easy to use. SciPy provides
MATLABMATLABMATLAB like features and can be used for processing
multidimensional arrays. Figure 1.1 summarizes the above discussion.
FIGURE 1.1 Some of the applications of Python
This chapter introduces the Python programming language. The chapter has been
organized as follows. Section 1.2 discusses the features of Python, Section 1.3
discusses the paradigms and Section 1.4 discusses the development and uses.
The installation of Anaconda has been introduced in Section 1.5. The last section
concludes the chapter.
1.2 FEATURES OF PYTHON

As stated earlier, Python is a simple but powerful language. Python is portable. It
has built-in object types, many libraries and is free. This section briefly discusses
the features and strengths of Python.
1.2.1 Easy
Python is easy to learn and understand. As a matter of fact, if you are from a
programming background you will find it elegant and uncluttered. The removal
of braces and parentheses makes the code short and sweet. Also, some of the
tasks in Python are pretty easy. For example, swapping numbers in Python is as
easy as writing (a, b)= (b, a).
It may also be stated here that learning something new is an involved and
intricate task. However, the simplicity of Python makes it almost a cake walk.
Moreover, learning advanced features in Python is a bit intricate, but is worth the
effort. It is also easy to understand a project written in Python. The code, in
Python, is concise and effective and therefore understandable and manageable.
1.2.2 Type and Run
In most projects, testing something new requires scores of changes and therefore
recompilations and re-runs. This makes testing of code a difficult and time
consuming task. In Python, a code can be run easily. As a matter of fact, we run
scripts in Python.
As we will see later in this chapter, Python also provides the user with an
interactive environment, in which one can run independent commands.
1.2.3 Syntax
The syntax of Python is easy; this makes the learning and understanding process
easy. According to most of authors, the three main features which make Python
attractive are that it’s simple, small, and flexible.
1.2.4 Mixing
If one is working on a big project, with perhaps a large team, it might be the case
that some of the team members are good in other programming languages. This
may lead to some of the modules in some other languages wanting to be
embedded with the core Python code. Python allows and even supports this.
1.2.5 Dynamic Typing
Python has its own way of managing memory associated with objects. When an
object is created in Python, memory is dynamically allocated to it. When the life
cycle of the object ends, the memory is taken back from it. This memory
management of Python makes the programs more efficient.
1.2.6 Built in Object Types
As we will see in the next chapter Python has built in object types. This makes
the task to be accomplished easy and manageable. Moreover, the issues related
to these objects are beautifully handled by the language.
1.2.7 Numerous Libraries and Tools
In Python, the task to be accomplished becomes easy—really easy. This is

because most of the common tasks (as a matter of fact, not so common tasks too)
have already been handled in Python. For example, Python has libraries which
help users to develop GUI’s, write mobile applications, incorporate security
features and even read MRI’s. As we will see in the following chapters, the
libraries and supporting tools make even the intricate tasks like pattern
recognition easy.
1.2.8 Portable
A program written in Python can run in almost every known platform, be it

Windows, Linux, or Mac. It may also be stated here that Python is written in C.
1.2.9 Free
Python is not propriety software. One can download Python compilers from
among the various available choices. Moreover, there are no known legal issues
involved in the distribution of the code developed in Python.
1.3 THE PARADIGMS
1.3.1 Procedural
In a procedural language, a program is actually a set of statements which execute

sequentially. The only option a program has, in terms of manageability, is
dividing the program into small modules. “C,” for example, is a procedural
language. Python supports procedural programming. The first section of this
book deals with procedural programming.
1.3.2 Object-Oriented
This type of language primarily focuses on the instance of a class. The instance
of a class is called an object. A class is a real or a virtual entity that has an
importance to the problem at hand, and has sharp physical boundaries. For
example in a program that deals with student management, “student” can be a
class. Its instances are made and the task at hand can be accomplished by
communicating via methods. Python is object-oriented. Section 2 of this book
deals with the object-oriented programming.
1.3.3 Functional
Python also supports functional programming. Moreover, Python supports

immutable data, tail optimization, etc. This must be music to the ears for those
from a functional programming background. Here it may be stated that
functional programming is beyond the scope of this book. However, some of the
above features would be discussed in the chapters that follow.
So Python is a procedural, object-oriented and functional language.
1.4 CHRONOLOGY AND USES

Having seen the features, let us now move onto the chronology and uses of
Python. This section briefly discusses the development and uses of Python and
will motivate the reader to bind with the language.
1.4.1 Chronology
Python is written in C. It was developed by Guido Van Rossum, who is now the
Benevolent Director for Life of Python. The reader is expected to take note of
the fact that Python has got nothing to do with pythons or snakes. The name of
the language comes from the show “Monty Python’s Flying Circus,” which was
one of the favorite shows of the developer, Guido van Rossum. Many people
attribute the fun part of the language to the inspiration.
Python is easy to learn as the core of the language is pretty concise. The
simplicity of Python can also be attributed to the desire of the developers to
make a language that was very simple, easy to learn but quite powerful.
The continuous betterment of the language has been possible because of a

dedicated group of people, committed to supporting the cause of providing the
world with an easy yet powerful language. The growth of the language has given
rise to the creation of many interest groups and forums for Python. A change in
the language can be brought about by what is generally referred to as the PEP
(Python Enhancement Project). The PSF (Python Software Foundation) takes
care of this.
1.4.2 Uses
Python is being used to accomplish many tasks, the most important of which are
as follows:
Graphical User Interface (GUI) development

Scripting web pages
Database programming
Prototyping
Gaming
Component based programming
If you are working in Unix or Linux, you don’t need to install Python. This is
because in Unix and Linux systems, Python is generally pre-installed. However,
if you work in Windows or Mac then you need to download Python. Once you
have decided to download Python, look for its latest version. The reader is
requested to ensure that the version he/she intends to download is not an alpha or
a beta version. Reference 1 at the end of the book gives a brief overview of
distinctions between two of the most famous versions. The next section briefly
discusses the steps for downloading Anaconda, an open source distribution
software.
Many development environments are available for Python. Some of them are as
follows:
1. PyDev with Eclipse

2. Emacs
3. Vim
4. TextMate
5. Gedit
6. Idle
7. PIDA (Linux)(VIM based)
8. NotePad++ (Windows)
9. BlueFish (Linux)
There are some more options available. However, this book uses IDLE and
Anaconda. The next section presents the steps involved in the installation of
Anaconda.
1.5 INSTALLATION OF ANACONDA

In order to install Anaconda, go to https://docs.continuum.io/anaconda/install
and select the installer (Windows or Mac OS or Linux). This section presents the
steps involved in the installation of Anaconda on the Windows Operating
System.
First of all, one must choose the installer (32 bit or 64 bit). In order to do so,
click on the selected installer and download the .exe file. The installer will ask
you to install it on the default location. You can provide a location that does not
contain any spaces or Unicode characters. It may happen that during the
installation you might have to disable your anti-virus software. Figures 1.2(a) to
1.2(g) take the reader through the steps of installation.
FIGURE 1.2(a) The welcome screen of the installer, which asks the user to close all running
applications and then click Next
FIGURE 1.2(b) The license agreement to install Anaconda3 4.3.0 (32 bit)
FIGURE 1.2(c) In the third step, the user is required to choose whether he wants to install
Anaconda for a single user or for all the users
FIGURE 1.2(d) The user then needs to select the folder in which it will install
FIGURE 1.2(e) The user then must decide whether he wants to add Anaconda to path
environment variable and whether to register Anaconda as the default Python 3.6
The installation then starts. After installation, the following screen will appear:
FIGURE 1.2(f) When the installation is complete, this screen appears

FIGURE 1.2(g) You can also share your notebooks on cloud
Once Anaconda is installed, you can open Anaconda and run your scripts. Figure
1.3 shows the Anaconda navigator. From the various options available you can
choose the appropriate option for you. For example, you can open the
QTConsole and run the commands/ scripts. Figure 1.4 shows the snapshot of
QTConsole. The commands written may appear gibberish at this point, but will
become clear in the chapters that follow.
FIGURE 1.3 The Anaconda navigator

FIGURE 1.4 The QtConsole
1.6 CONCLUSION
Before proceeding any further, the reader must take note of the fact that some
things in Python are different when compared to any other language. The
following points must be noted to avoid any confusion.
In Python, statements do not end with any special characters. Python

considers the newline character as an indication of the fact that the
statement has ended. If a statement is to span more than a single line, the
next line must be preceded with a (\).
In Python, indentation is used to detect the presence of loops. The loops in
Python do not began or end with delimiters or keywords.
A file in Python is generally saved with a .py extension.
The shell can be used as a handy calculator.
The type of a variable need not to be mentioned in a program.
Choice at every step is good but can also be intimidating. As stated earlier,
Python’s core is small and therefore it is easy to learn. Moreover, there are some
things like (if/else), loops and exception handling which are used in almost all
the programs.
The chapter introduces Python and discusses the features of Python. One must
appreciate the fact that Python supports all three paradigms: procedural, object-
oriented, and functional. This chapter also paves the way for the topics presented
in the following chapters. It may also be stated that the codes presented in this
book will run on versions 3.X.
GLOSSARY
PEP: Python Enhancement Project
PSF: Python Software Foundation
POINTS TO REMEMBER
Python is a strong procedural, object-oriented, functional language crafted
in late 1980s by Guido Van Rossum.
Python is open source.
The applications of Python include software development, web
development, desktop GUI development, education and scientific
applications.
Python is popular due to its simplicity and robustness.
It is easy to interface with C++ and Java.
SciPy is used for engineering and mathematics, IPython for parallel
computing etc., Scions is used for build control.
The various development environments for Python are PyDev with Eclipse,
Emacs, Vim, TextMate, Gedit, Idle, PIDA (Linux)(VIM Based), NotePad++
(Windows), and BlueFish (Linux).
RESOURCES
To download Python, visit www.python.org
The documentation is available at www.python.org/doc/
EXERCISES
MULTIPLE CHOICE QUESTIONS
1. Python can subclass a class made in

(a) Python only
(b) Python, C++
(c) Python, C++, C#, Java
(d) None of the above
2. Who created Python?

(a) Monty Python
(b) Guido Van Rossum
(c) Dennis Richie
3. Monty Python was

(a) Creator of Python Programming Language
(b) British Comedy Group
(c) American Band
(d) Brother of Dosey Howser
4. In Python, libraries and tools

(a) Not supported
(b) Supported but not encouraged
(c) Supported and encouraged
(d) Supported (only that of PSF’s)
5. Python has
(a) Built in object types
(b) Data types
(c) Both
6. Python is a
(a) Procedural language
(b) object-oriented Language
(c) Fictional
(d) All of the above
7. There is no data type, so a code in Python is applicable to whole range of

Objects. This is called
(a) Dynamic Binding
(b) Dynamic Typing
(c) Dynamic Leadership
8. Which of the following is automatic memory management?

(a) Automatically assigning memory to objects
(b) Taking back the memory at the end of life cycle
(c) Both
9. PEP is
(a) Python Ending Procedure
(b) Python Enhancement proposal
(c) Python Endearment Project
(d) none of the above
10. PSF is
(a) Python Software Foundation
(b) Python Selection Function
(c) Python segregation function
11. What can be done in Python
(a) GUI
(b) Internet scripting
(c) Games
12. What can be done using Python?

(a) System programming
(b) Component based programming
(c) Scientific programming
13. Python is used in

(a) Google
(b) Raspberry Pi
(c) Bit Torrent
14. Python is used in

(a) App Engine
(b) YouTube sharing
(c) Real time programming
15. Which is faster?

(a) PyPy
(b) IDLE
(c) Both are equally good
(d) depends on the task
THEORY
1. Write the names of three projects which are using Python.
2. Explain a few applications of Python.
3. What type of language is Python? (Procedural, object-oriented or functional)
4. What is PEP?
5. What is PSF?
6. Who manages Python?
7. Is Python open source or proprietary?
8. What languages can be supported by Python?
9. Explain the chronology of the development of Python.
10. Name a few editors for Python.
11. What are the features of Python?
12. What is the advantage of using Python over other languages?
13. What is Dynamic Typing?
14. Does Python have data types?
15. How is Python different from Java?

CHAPTER
2
PYTHON OBJECTS
• Understand the meaning and importance of variables, operators, keywords, and objects
• Use numbers and fractions in a program
• Appreciate the importance of strings
• Understand slicing and indexing in strings
• Use of lists and tuples
• Understand the importance of tuples
2.1 INTRODUCTION
To be able to write a program in Python the programmer can use Anaconda, the
installation of which was described in the previous chapter—or you can use
IDLE, which can be downloaded from the reference given at the end of the
Chapter 1. IDLE has an editor specially designed for writing a Python program.
As stated earlier Python is an interpreted language, so one need not to compile

every piece of code. The programmer can just write the command and see the
output at the command prompt. For example, when writing 2+3 on the command
line we get
>>2+3
5
As a matter of fact you can add, subtract, multiply, divide and perform
exponentiation in the command line. Multiplication can be done using the *
operator, the division can be performed using the / operator, the exponentiation
can be done using the ** operator and the modulo can be found using the %
operator. The modulo operator finds the remained if the first number is greater
than the other, otherwise it returns the first number as the output. The results of
the operations have been demonstrated as follows:
>>> 2*3
6
>>> 2/3
0.6666666666666666
>>> 2**3
8
>>> 2%3
2
>>> 3%2
1
>>>
In the above case, the Python interpreter is used to execute the commands. This
is referred to as a script mode. This mode works with small codes. Though
simple commands can be executed on the command line, the complex programs
can be written in a file. A file can be created as follows:
Step 1. Go to FILE NEW
Step 2. Save the file as calc.py
Step 3. Write the following code in the file

print(2+3)
print(2*3)
print(2**3)
print(2/3)
print(2%3)
print(3/2)
Step 4. Go to debug and run the program. The following output will be
displayed.
>>>
============ RUN C:/Python/Chapter 2/calc.py ============
5
6
8
0.6666666666666666
2
1.5
>>>
Conversely, the script can be executed by writing Python calc.py on the

command prompt. In order to exit IDLE go to FILE->EXIT or write the exit()
function at the command prompt.
In order to store values, we need variables. Python empowers the user to

manipulate variables. These variables help us to use the values later. As a matter
of fact, everything in Python is an object. This chapter focuses on objects. Each
object has identity, a type, and a value (given by the user / or a default value).
The identity, in Python, refers to the address and does not change. The type can
be any of the following.
None: This represents the absence of a value.
Numbers: Python has three types of numbers:
Integer: It does not have any fractional part

Floating Point: It can store number with a fractional part
Complex: It can store real and imaginary parts
Sequences: These are ordered collections of elements. There are three types of
sequences in Python:
String
Tuples
Lists
These types have been discussed in the sections that follow.
Sets: This is an un-ordered collection of elements.
Keywords: These are words having special meanings and are understood by the
interpreter. For example, and, del, from, not, while, as, elif, global,
else, if, pass, Yield, break, except, import, class, raise,
continue, finally, return, def, for, and try are some of the keywords
which have been extensively used in the book. For a complete list of keywords,
the reader may refer to the Appendix.
Operators: These are special symbols which help the user to carry out
operations like addition, subtraction, etc. Python provides following type of
operators:
Arithmetic operators: +, –, *, /, %, ** and //.

Assignment operators: =, + =, – =, *=, /=, %=, **= and //=
Logical operators: or, and, and not
Relational operators: <, <=, >, >=, != or < > and ==.
This chapter deals with the basic data types in Python and their uses. The chapter
has been organized as follows: Section 2 of this chapter deals with the
introduction to programming in Python and basic data types, and Section 3 deals
with strings. Section 4 deals with lists and tuples. The last section of this chapter
concludes the chapter. The readers are advised to go through the references at the
end of this book for comprehensive coverage of the topic.
2.2 BASIC DATA TYPES REVISITED

The importance of data types has already been discussed. There is another
reason to understand and to be able to deal with built-in data types, which is that
they generally are an intrinsic part of the bigger types which can be developed
by the user.
The data types provided by Python are not only powerful but also can be nested
within others. In the following discussion the concept of nested lists has been
presented, which is basically a list within a list. The power of data types can be
gauged by the fact that Python provides the user with dictionaries, which makes
mapping easy and efficient.
Numbers are the simplest data types. Numbers comprise of integers, floats,
decimals, and complexes in Python. The type of numbers and their explanations
have been summarized in Table 2.1. The operators supported by numbers have
been presented in Table 2.2.
Table 2.1 Numbers
Numbers Explanation
Integers Which do not have any fractional part
Floating point numbers That do have a fractional part
Complex numbers The numbers having a real and an imaginary part
Decimal Those having fixed precision
Rational Those having a numerator and a denominator
Sets Abstraction of a mathematical set
Table 2.2 Operators supported in numbers
+ Addition
– Subtraction
* Multiplication
** Power
% Modulo
In addition to the above, Python is practically free from the problems of C and
C++ and can calculate very, very large integers. Let us now have a look at how
to use these operators. For example if one needs to calculate the square root of a
number, then importing math and using math.sqrt() is a solution. Some of the
most important functions have been explained in the following sneak peek.
Sneak Peek
1. Ceil: The ceiling of a given number is the nearest integer greater than or
equal to that number. For example, the ceiling of 2.678 is 3.
>>> import math
>>>math.ceil(2.678)
3
That of 2 is 2.
>>>math.ceil(2)
2
>>>
2. Copy sign: The sign of the second argument is returned along with the result
on the execution of this function.
math.copysign(x, y)
Return x with the sign of y.
On a platform that supports signed zeros, copy sign (1.0, – 0.0) returns –1.0.
3. Fabs: The absolute value of a number is its positive value; that is if the
number is positive then the number itself is returned. If, on the other hand,
the number is negative then it is multiplied by –1 and returned.
In Python, this task is accomplished with the function fabs (x).

The fabs(x) returns the absolute value of x.
>>>math.fabs(-2.45)
2.45
>>>math.fabs(x)
Return the absolute value of x.
4. Factorial: The factorial of a number x is defined as the continued product of
the numbers from 1 to that value. That is:
Factorial(x) = 1 × 2 × 3 × … × n.
In Python, the task can be accomplished by the factorial function math.
factorial(x).
It returns the factorial of the number x. Also if the given number is not an
integer or is negative, then an exception is raised.
5. Floor: The floor of a given number is the nearest integer smaller than or
equal to that number. For example the floor of 2.678 is 2 and that of 2 is also
2.
>>> import math
>>>math.floor(2.678)
2
>>>math.floor(2)
2
>>>
2.2.1 Fractions
Python also provides the programmer the liberty to deal with fractions. The use
of fractions and decimals has been shown in the following listing.
Listing
from fractions import Fraction
print(Fraction(128, -26))
print(Fraction(256))
print(Fraction())
print(Fraction('2/5'))
print(Fraction(' -5/7'))
print(Fraction('2.675438 '))
print(Fraction('-32.75'))
print(Fraction('5e-3'))
print(Fraction(7.85))
print(Fraction(1.1))
print(Fraction(2476979795053773, 2251799813685248))
from decimal import Decimal
print(Fraction(Decimal('1.1')))
>>>
Output
========== RUN C:/Python/Chapter 2/Fraction.py ==========
-64/13
256
0
2/5
-5/7
1337719/500000
-131/4
1/200
4419157134357299/562949953421312
2476979795053773/2251799813685248
2476979795053773/2251799813685248
11/10
>>>
2.3 STRINGS
In Python a string is a predefined object which contains characters. The string in
Python is non-mutable; that is, once defined the value of a string cannot be
changed. However, as we proceed further, the exceptions to the above premise
will be discussed. To begin with, let us consider a string containing value
“Harsh,” that is:
name = 'Harsh'
The value of this string can be displayed simply by typing the name of the object
(name in this case) into the command prompt.
>>>name
Harsh
The value can also be printed by using the print function, explained previously.
print(name)
The value at a particular location of a string can be displayed using indexing.

The syntax of the above is as follows.
<name of the String>[index]
It may be stated here that the index of the first location is 0. So, name[0] would
print the first letter of the string, which is “H.”
print(name[0])
H
Negative indexing in a string refers to the character present at the nth position
beginning from the end. In the above case, name[-2] would generate “s.”
print(name[-2])
s
The length of a string can be found by calling the len function. len(str) returns
the length of the string “str.” For example, len(name) would return 5, as
'harsh' has 5 characters.
The last character of a given string can also be printed using the following.
print(name[len(name)-1])
The + operator concatenates, in the case of a string. For example “harsh” +

“arsh” would return “Harsharsh,” that is
name = name + 'arsh'

print(name)
Harsharsh
After concatenation, if the first and the second last characters are to be printed
then the following can be used.
print(name[0])
print(name[-2])
print(name[len(name)-1→2])
H
S
s
The * operator, of string, concatenates a given string the number of times, given
as the first argument. For example, 3*name would return “harsharshharsharsh.”
The complete script as follows:
Listing
name = 'Harsh'
print(name)
print(name[0])
print(name[-2])
name = name + 'arsh'
print(name)
print(name[0])
print(name[-2])
>>>
Output
=========== RUN C:/Python/Chapter 2/String.py ===========
Harsh
H
s
h
Harsharsh
H
s
h
>>>
Slicing: Slicing, in strings, refers to removing some part of a string. For

example:
>>>name = 'Sonam'
>>>name
'Sonam'
Here, if we intend to extract the portion after the first letter we can write [1:].
>>> name1=name[1:]
>>> name1
'onam'
In the same way the portion of the string after the first two letters can be
extracted as follows.
>>>name = name[2:]
>>>name
'nam'
Now, we modify the string by adding “man man”
>>>name = “man”+name
>>>name
'mannam'
It may be noted that the last two characters cannot be removed in the same way
as the first two. Observe the following output in order to understand the concept.
>>>name = name[:2]
>>>name
'ma'
>>>name = “man manam”
In order to accomplish the above task, negative indexing ought to be used.
>>>name
'manmanam'
>>> name2 = name[:-2]
>>> name2
'man man'
>>>
Immutability of Strings
It may be noted that when we write
name = 'Hello' + name
we don’t actually change the string; as a matter of fact we create a new string
having the value 'Hello' concatenated with the value stored in name. The
concept can be understood by the fact that when we try to change the value of a
particular character in a string, an error crops up.
>>>name='Anupam'
>>>name
'Anupam'
>>>name[2]='p'
Traceback (most recent call last):
File “<pyshell#17>”, line 1, in <module>
name[2]='p'
TypeError: 'str' object does not support item assignment
>>>
2.4 LISTS AND TUPLES
2.4.1 List
A list, in Python, is a collection of objects. As per Lutz “It is the most general
sequence provided by the language.” Unlike strings, lists are mutable. That is, an
element at a particular position can be changed in a list. A list is useful in dealing
with homogeneous and heterogeneous sequences.
A list can be one of the following:
A list can be a collection of similar elements (homogeneous), for example

[1, 2, 3]
It can also contain different elements (heterogeneous), like [1, “abc,” 2.4]
A list can also be empty ([])
A list can also contain a list (discussed in Chapter 4, of this book)
For example, the following list of authors has elements “Harsh Bhasin,” “Mark
Lutz,” and “Shiv.” The list can be printed using the usual print function. In the
following example, the second list in the following listing contains a number, a
string, a float, and a string. “list 3” is a null list and list-of-list contains list as its
elements.
Listing
authors = ['Harsh Bhasin', 'Mark Lutz', 'Shiv']
print(authors)
combined =[1, 'Harsh', 23.4, 'a']
print(combined)
list3= []
print(list3)
listoflist = [1, [1,2], 3]
print(listoflist)
>>>
Output
============ RUN C:/Python/Chapter 2/Lists.py ===========
['Harsh bhasin', 'Mark Lutz', 'Shiv']
[1, 'Harsh', 23.4, 'a']
[]
[1, [1, 2], 3]
>>>
An element of a list can be accessed by indexing; for example if list 1 contains

[1, 2, 3], then list 1[1] contains “2” and list 1[-1] contains “3.”
Listing
list1 = [1, 2, 3]
print(list1[1])
print(list1[-1])
>>>
Output
=========== RUN C:/Python/Chapter 2/list2.py ============
2
3
>>>
A list can also contain list(s). The topic has been discussed in Chapter 4. Lists
also support slicing.
2.4.2 Tuples
A tuple contains elements which can be treated individually or as a group. A

tuple (say (x, y)) can be printed using the standard print( ) function. The elements
of a tuple can be accessed by assigning it to a tuple, as shown in the following
listing. A tuple may also contain heterogeneous elements. For example, in the
following listing, tup2 and tup3 contain a string and an integer.
Listing
tup1= (2, 3)
print(tup1)
(a, b) = tup1
print('The first element is ',a)
print('The second element is ',b)
tup2=(101, 'Hari')
tup3=(102,'Shiv')
(code1, name1)=tup1
(code2, name2)=tup2
print('The code of ', name1,' is ',code1,'\nThe code of
',name2, ' is ',code2)
>>>
Output
=========== RUN C:/Python/Chapter 2/tuple.py ============
(2, 3)
The first element is 2
The second element is 3
The code of 3 is 2
The code of Hari is 101
>>>
Tuples are extremely useful in operations like swapping etc. Swapping in Python
is as simple as assigning (a, b) to (b, a). The program for swapping two numbers
using tuples has been given as follows.
Illustration 2.1: Write a program to swap two numbers using tuples.
Solution:
print('Enter the first number\t:')

num1= int(input())
print('Enter the second number\t:')
num2= int(input())
print('\nThe numbers entered are ',num1,' & ', num2)
(num1, num2) = (num2, num1)
print('\nThe numbers now are ',num1,' & ', num2)
>>>
Output
============ RUN C:/Python/Chapter 2/swap.py ============
Enter the first number :
2
Enter the second number :
3
The numbers entered are 2& 3
The numbers now are 3& 2
>>>
2.4.3 Features of Tuples
Tuples are immutable—an element of a tuple cannot be assigned a different

value once it has been set. For example,
tup1 = (2, 3)
tup1[1] = 4
would raise an exception.
The “+” operator in a tuple concatenates two tuples. For example,

>>> tup1= (1,2)
>>> tup2=(3,4)
>>> tup3= tup1+tup2
>>> tup3
(1, 2, 3, 4)
>>>
2.5 CONCLUSION
In a program, instructions are given to a computer to perform a task. To be able
to do so, the operators operate on what are referred to as “objects.” This chapter
explains the various types of objects in Python and gives a brief overview of the
operators that act upon them. The objects can be built in or user defined. As a
matter of fact, everything that will be operated upon is an object.
The first section of this chapter describes various built-in objects in Python. The
readers familiar with “C” must have an idea as to what a procedural language is.
In “C,” for example, a program is divided into manageable modules, each of
which performs a particular task. The division of bigger tasks into smaller parts
makes parts manageable and the tracking of bugs easy. There are many more
advantages of using modules, some of which have been stated in Chapter 1.
These modules contain a set of statements, which are equivalent to instructions
(or no instruction, e.g. in case of a comment). The statements may contain
expressions, in which objects are operated upon by operators. As stated earlier,
Python gives its user the liberty to define their own objects. This will be dealt
with in the chapter on classes and objects. This chapter focuses on the built in
objects.
In C (or for that matter C++), one needs to be careful not only about the built-in
type used but also about the issues relating to the allocation of memory and data
structures. However, Python spares the user of these problems and can therefore
focus on the task at hand. The use of built-in data types makes things easy and
efficient.
GLOSSARY
None: This represents the absence of value.
Numbers: Python has three types of numbers: integers, floating point, complex.
Sequences: These are ordered collections of elements. There are three types of
sequences in Python:
String
Tuples
Lists
POINTS TO REMEMBER
In order to store values, we need variables.
Everything in Python is an object.
Each object has identity, a type, and a value.
EXERCISES

1. >>> a = 5
>>> a + 2.7
>>> a
(a) 7.7
(b) 7
(c) None of the above
(d) An exception is raised
2. >>> a = 5
>>> b = 2
>>> a/b
(a) 2
(b) 2.5
(c) 3
3. >>> a = 5
>>> b = 2
>>> c = float (a)/b
>>> c
(a) 2
(b) 2.5
(c) 3
4. >>> a = 2
>>> b = 'A'
>>> c = a + b
>>> c
(a) 67
(b) 60
5. >>> a = 'A'
>>> 2*A
(a) ‘AA’
(b) 2A
(c) A2
6. >>> a = 'A'
>>> b = 'B'
>>> a + b
(a) A + B
(b) AB
(c) BA
7. >>> (a, b) = (2, 5)

>>> (a, b) = (b, a)
>>> (a, b)
(a) (2, 5)
(b) (5, 2)
(c) (5, 5)
8. >>> a = 5
>>> b = 2
>>> a = a + b
>>> b = a - b
>>> a = a - b
>>> a
(a) 5
(b) 2
9. >>> a = 5
>>> b * b = a
>>> b
(a) 2.7
(b) 25
10. >>> (a, b) = (2, 3)

>>> (c, d) = (4, 5)
>>> (a, b) + (c, d)
(a) (6, 8)
(b) (2, 3, 4, 5)
(c) (8, 6)
11. In the above question what would (a, b) – (c, d) generate

(a) (6, 8)
(b) (2, 3, 4, 5)
(c) (8, 6)
12. In the above question what would (a, b) * (c, d) generate

(a) (6, 8)
(b) (2, 3, 4, 5)
(c) (8, 6)
13. >>> a = 'harsh'

>>> b = a[1: len(a)]
>>> b
(a) arsh
(b) hars
(c) harsh
14. >>>a = 'harsh'

>>>b = [-3, len (a)]
(a) rsh
(b) arsh
(c) harsh
15. >>>b
>>>a = 'tar'
>>>b = 'rat'
>>>2*(a + b) is
(a) tarrattarrat
(b) rattarrattar
(c) tarratrattar
PROGRAMS
1. Write a program to swap two numbers.
2. Ask the user to enter the coordinates of a point and find the distance of the
point from the origin.
3. Ask the user to enter two points (x and y coordinates) and find the distance
between them.
4. Ask the user to enter three points and find whether they are collinear.
5. In the above question, if the points are not collinear then find the type of
triangle formed by them (equilateral, isosceles or scalene).
6. In the above question, check if the triangle is right angled.
7. In question number 4, find the angles of the triangle.
8. Ask the user to enter two points and find if they are at equal distances from
the origin.
9. In question number 8, find the angle between the line joining the points and
the origin.
10. Ask the user to enter 4 points and arrange them in order of their distances
from the origin.
11. In question 10, arrange the above points in order of their x co-ordinates.
CHAPTER
3
CONDITIONAL STATEMENTS
• Use conditional statements in programs
• Appreciate the importance of the if-else construct
• Use the if-elif-else ladder
• Use the ternary operator
• Understand the importance of & and |
• Handle conditional statements using the get construct
3.1 INTRODUCTION
The preceding chapters presented the basic data types and simple statements in
Python. The concepts studied so far are good for the execution of a program
which has no branches. However, a programmer would seldom find a problem
solving approach devoid of branches.
Before proceeding any further let us spare some time contemplating life. Can
you move forward in life without making decisions? The answer is NO. In the
same way the problem solving approach would not yield results until the power
of decision making is incorporated. This is the reason one must understand the
implementation of decision making and looping. This chapter describes the first
concept. This is needed to craft a program which has branches. Decision making
empowers us to change the control-flow of the program. In C, C++, Java, C#,
etc., there are two major ways to accomplish the above task. One is the ‘if’
construct and the other is ‘switch’. The ‘if’ block in a program is executed if
the ‘test’ condition is true otherwise it is not executed. Switch is used to
implement a scenario in which there are many ‘test’ conditions, and the
corresponding block executes in case a particular test condition is true.
The chapter introduces the concept of conditional statements, compound

statements, the if-elif ladder and finally the get statement. The chapter
assumes importance as conditional statements are used in every aspect of
programming, be it client side development, web development, or mobile
application development.
The chapter has been organized as follows. The second section introduces the
‘if’ construct. Section 3.3 introduces ‘if-elif’ ladder. Section 3.4 discusses the
use of logic operators. Section 3.5 introduces ternary operators. Section 3.6
presents the get statement and the last section concludes the chapter. The reader
is advised to go through the basic data types before proceeding further.
3.2 IF, IF-ELSE, AND IF-ELIF-ELSE CONSTRUCTS
Implementing decision making gives the power to incorporate branching in a

program. As stated earlier, a program is a set of instructions given to a computer.
The instructions are given to accomplish a task and any task requires making
decisions. So, conditional statements form an integral part of programming. The
syntax of the construct is as follows:
General Format
1. if
if <test condition>:
<block if the test condition is true>
2. if-else
else:
<block if the test condition is not true>
...
3. If else ladder (discussed in the next section)
elif <test 2>:
<second block>
elif <test 3>:
<third block>
else:
Note that indentation is important, as Python recognizes a block through

indentation. So, make sure that the 'if (<condition>):' is followed by a
block, each statement of which is at the same alignment. In order to understand
the concept, let us consider a simple example. A student generally clears a
university exam in India if he scores more than 40%. In order to implement the
logic, the user is asked to enter the value of the percentage. If the percentage
entered is more than 40 then “exam cleared” is printed, otherwise “failed” is
printed. The situation has been depicted in the following figure (Figure 3.1).
FIGURE 3.1 Flow chart for example 1
Illustration 3.1: Ask the user to enter the marks of a student in a subject. If the
marks entered are greater than 40 then print “pass,” if they are lower print “fail.”
Program
>>>a = input("Enter marks : ")
if int(a)> 40:
print('Pass')
else:
print('Fail')
...
Output 1
Enter Marks : 50
Pass
Output 2
Enter Marks : 30
Fail
Let us have a look at another example. In the problem, the user is asked to enter
a three digit number to find the number obtained by reversing the order of the
digits of the number; then find the sum of the number and that obtained by
reversing the order of the digits and finally, find whether this sum contains any
digit in the original number. In order to accomplish the task, the following steps
(presented in Illustration 3.2) must be carried out.
Illustration 3.2: Ask the user to enter a three digit number. Call it 'num'. Find
the number obtained by reversing the order of the digits. Find the sum of the
given number and that obtained by reversing the order of the digits. Finally, find
if any digit in the sum obtained is the same as that in the original number.
Solution:
The problem can be solved as follows:
When the user enters a number, check whether it is between 100 and 999,
both inclusive.
Find the digits at unit’s, ten’s and hundred’s place. Call them 'u', 't' and
'h' respectively.
Find the number obtained by reversing the order of the digits (say, ‘rev’)
using the following formula.
Number obtained by reversing the order of the digits, rev = h + t × 10 + u ×
100
Find the sum of the two numbers.
Sum = rev + num
The sum may be a three digit or a four digit number. In any case, find the
digits of this sum. Call them 'u1', 't1', 'h1' and 'th1' (if required).
Set 'flag=0'.
Check the following condition. If any one is true, set the value of flag to 1.
If “sum” is a three digit number
u = = u1
u = = t1
u = = h1
t = = u1
t = = t1
t = = h1
h = = u1
h = = t1
h = = h1
If “sum” is a four digit number the above conditions need to be checked
along with the following conditions:
u = = th1
h = = th1
t = = th1
The above conditions would henceforth be referred to as “set 1.” If the
value of “flag” is 1, then print 'true' else print 'false'.
The above process has been depicted in the following figure (Figure 3.2).
FIGURE 3.2 Flow chart for Illustration 2
Program
Output: First run
>>>
========= RUN C:/Python/Conditional/Problem 2.py =========
Enter a three digit number :4
You have not entered a number between 100 and 999
>>>
Output: Second run

>>>
o : 3 t : 4 h : 3
Number obtained by reversing the order of the digits : 343
No digit of the sum is same as the original number
>>>
Output: Third run

>>>
Enter a three digit number : 435
o : 5 t : 3 h : 4
No digit of the sum is same as the original number
>>>
Output: Fourth run

>>>
o : 1 t : 2 h : 1
Sum of the number and that obtained by reversing the order
of digits : 242
o1 : 2 t1 : 4 h1 : 2
Condition true
>>>
.
Tip
One must be careful regarding the indentation, failing which the program would
not compile. The indentation decides the beginning and ending of a particular
block in Python. It is advisable not to use a combination of spaces and tabs in
indentation. Many versions of Python may treat this as a syntax error.
The if-elif ladder can also be implemented using the get statement, explained
later in the chapter. The important points regarding the conditional statements in
Python are as follows:
The if <test> is followed by a colon.

There is no need of parentheses for this test condition. Though enclosing
test in parentheses will not result in an error.
The nested blocks in Python are determined by indentation. Therefore,
proper indentation in Python is essential. As a matter of fact, an inconsistent
indentation or no indentation will result in errors.
An if can have any number of if's nested within.
The test condition in if must result in a True or a False.
Illustration 3.3: Write a program to find the greatest of the three numbers
entered by the user.
Solution: First of all, three variables (say num1, num2, and num3) are needed.
These variables will get their values from the user. This input will be followed
by the condition checking as depicted in the following program. Finally, the
greatest number will be displayed. The listing is given as follows:
Program
3.3 THE IF-ELIF-ELSE LADDER

If there are multiple conditions and the outcomes decide the action, then an if-
elif- else ladder can be used. This section discusses the construct and presents
the concept using relevant examples. The syntax of this construct is as follows:
Syntax
if <test condition 1>:
# The task to be performed if the condition 1 is true
elif <test condition 2>:
elif <test condition 3>:
else:
# The task to be performed if none of the above condition
is true
The flow of the program can be managed using the above construct. Figure 3.3
shows the diagram depicting the flow of the program which uses the above
constructs.
In the figure, the left edge depicts the scenario where the condition C is true and
the right edge depicts the scenario where the condition is false. In the second
graph, conditions C1, C2, C3, and C4 lead to different paths [Programming in
C#, Harsh Bhasin, 2014].
FIGURE 3.3 The flow graph of if and elif ladder
The following section has programs that depict the use of the elif ladder. It may
be noted that if there are multiple else statements, then the second else is taken
along with the nearest if.
3.4 LOGICAL OPERATORS

In many cases the execution of a block depends on the truth value of more than
one statement. In such cases the operators “and” (“&”) and “or” (“|”) come to
our rescue. The first ('and') is used when the output is 'true', when both the
conditions are 'true'. The second ('or') is used if the output is 'true', if any
of the conditions are 'true'.
The truth table of 'and' and 'or' is given as follows. In the tables that follow
“T” stands for “true” and “F” stands for “false.”
Table 3.1 Truth table of a&b
a b a&b
t T T
t F F
F T F
F F F
Table 3.2 Truth table of a|b
a b a|b
t T T
t F T
F T T
F F F
The above statement helps the programmer to easily handle compound

statements. As an example, consider a program to find the greatest of the three
numbers entered by the user. The numbers entered by the user are (say) 'a', 'b',
and 'c', then 'a' is greatest if (a > b) and (a > c). This can be written as
follows:
if((a>b)&(a>c))
print('The value of a greatest')
In the same way, the condition of ‘b’ being greatest can be crafted. Another
example can be that of a triangle. If all the three sides of a triangle are equal,
then it is an equilateral triangle.
if((a==b)||(b==c)||(c==a))
//The triangle is equilateral;
3.5 THE TERNARY OPERATOR

The conditional statements explained in the above section are immensely
important to write any program that contains conditions. However, the code can
still be reduced further by using the ternary statements provided by Python. The
ternary operator performs the same task as the if-else construct. However, it has
the same disadvantage as in the case of C or C++. The problem is that each part
caters to a single statement. The syntax of the statement is given as follows.
Syntax
<Output variable> = <The result when the condition is true>
if <condition> else <The result when the condition is not
true>
For example, the conditional operator can be used to check which of the two
numbers entered by the user is greater.
great = a if (a>b) else b
Finding the greatest of the three given numbers is a bit intricate. The following
statement puts the greatest of the three numbers in “great.”
great = a if (a if (a > b) else c)) else(b if (b>c) else c))
The program that finds the greatest of the three numbers entered by the user
using a ternary operator is as follows.
Illustration 3.4: Find the greatest of three numbers entered by the user, using a
ternary operator.
Program
a = int(input('Enter the first number\t:'))
b = int(input('Enter the second number\t:'))
c = int(input('Enter the third number\t:'))
big = (a if (a>c) else c) if (a>b) else (b if (b>c) else c)
print('The greatest of the three numbers is '+str(big))
>>>
Output
========== RUN C:/Python/Conditional/big3.py ==========
Enter the first number 2
Enter the second number 3
Enter the third number 4
The greatest of the three numbers is 4
>>>
3.6 THE GET CONSTRUCT

In C or C++ (even in C# and Java) a switch is used in the case where different
conditions lead to different actions. This can also be done using the'if-elif'
ladder, as explained in the previous sections. However, the get construct greatly
eases this task in the case of dictionaries.
In the example that follows there are three conditions. However, in many
situations there are many more conditions. The contact can be used in such
cases. The syntax of the construct is as follows:
Syntax
<dictionary name>.get('<value to be searched>', 'default
value>')
Here, the expression results in some value. If the value is value 1, then block 1
is executed. If it is value 2, block 2 is executed, and so on. If the value of the
expression does not match any of the cases, then the statements in the default
block are executed. Illustration 5 demonstrates the use of the get construct.
Illustration 3.5: This illustration has a directory containing the names of books
and the corresponding year they were published. The statements that follow find
the year of publication for a given name. If the name is not found the string
(given as the second argument, in get) is displayed.
Program
hbbooks = {'programming in C#': 2014, 'Algorithms': 2015,
'Python': 2016}
print(hbbooks.get('Progarmming in C#', 'Bad Choice'))
print(hbbooks.get('Algorithms', 'Bad Choice'))
print(hbbooks.get('Python', 'Bad Choice'))
print(hbbooks.get('Theory Theory, all the way', 'Bad
Choice'))
Output
>>>
========== RUN C:/Python/Conditional/switch.py ==========
Bad Choice
2015
2016
Bad Choice
>>>
Note that in the first case the “P” of “Programming” is capital, hence “Bad
Choice” is displayed. In the second and the third cases, the get function is able to
find the requisite value. In the last case the value is not found and so the second
argument of the get function appears. Note that it is similar to the default of the
“C” type switch statement. The flow diagram given in Figure 3.4 shows a
program that has many branches.
FIGURE 3.4 A program having multiple conditional statements

Observation
In Python, dictionaries and lists form an integral part of the language basics. The
use of the get construct was not explained in Chapter 2 of this book, as it
implements the concept of conditional selection. It may be noted that this
construct greatly reduces the problems of dealing with the situations where
mapping is required and is therefore important.
3.7 EXAMPLES
The 'if' condition is also used for input validation. The process will be
explained in the following sections of this book. However, the idea has been
used in the example that follows. The program asks the user to enter a character
and checks whether its ASCII value is greater a certain value.
Illustration 3.6: Ask the user to enter a number and check whether its ASCII
value is greater than 80.
Program
inp = input('Enter a character :')
if ord(inp) > 80:
print('ASCII value is greater than 80')
else:
print('ASCII value is less than 80')
Output 1:
>>>Enter a character: A
ASCII value is less than 80
...
Output 2
>>>Enter a character: Z
ASCII value is greater than 80
>>>
The construct can also be used to find the value of a multi-valued function. For
example, consider the following function:
The following example asks the user to enter the value of x and calculates the
value of the function as per the given value of x.
Illustration 3.7: Implement the above function and find the values of the
function at x = 2 and x = 4.
Program
Output
========== RUN C:\Python\Conditional\func.py ==========
Enter the value of x :4
Value of function f(x) = 39
>>>
========== RUN C:\Python\Conditional\func.py ==========
Enter the value of x :1
Value of function f(x) = 4
>>>
The 'if-else' construct, as stated earlier, can be used to find the outcome based
on certain conditions. For example two lines are parallel if the ratio of the
coefficients of x’s is same as that of those of y’s.
For a1x + b1y + c1 = 0 and a2x + b2y + c2 = 0. Then the condition of lines being
parallel is:
The following program checks whether two lines are parallel or not.
Illustration 3.8: Ask the user to enter the coefficients of a1x + b1y + c1 = 0 and
a2x + b2y + c2 = 0 and find out whether the two lines depicted by the above
equations are parallel or not.
Program
Output
>>>
========== RUN C:\Python\Conditional\parallel.py ==========
Enter Coefficients of the first equation [a1x + b1y + c1 =
0]
Enter the value of a1: 2
Enter the value of b1: 3
Enter the value of c1: 4
Enter Coefficients of second equation [a2x + b2y + c2 = 0]

Lines are parallel
>>>
The above program can be extended to find whether the lines are intersecting or
overlapping: two lines intersect if the following condition is true.
a1x + b1y + c1 = 0 and a2x + b2y + c2 = 0. Then the lines intersect if:
And the two lines overlap if:
The following flow-chart shows the flow of control of the program (Figure 3.5).
FIGURE 3.5 Checking whether lines are parallel, overlapping, or if they intersect
The following program implements the above logic.
Illustration 3.9: Ask the user to enter the values of a1, a2, b1, b2, c1, and c2 and
find whether the lines are parallel, or if they overlap or intersect.
Program
Output
>>>
========== RUN C:/Python/Conditional/Lines.py ==========
Enter Coefficients of the first equation [a1x + b1y + c1 =
0]
Enter Coefficients of second equation [a2x + b2y + c2 = 0]
Lines intersect
>>>
3.8 CONCLUSION
As stated in the first chapter, we write a program with a purpose. The purpose to
be accomplished by a program generally requires making decisions. This
decision making capacity also empowers a programmer to write a code that
requires branching. Python greatly reduces unnecessary clutter when compared
to C or C++. In a Python code there is hardly a need for braces, or for that matter
handling obvious conditions. Python also provides us with a switch-like
construct to handle multiple conditions. This chapter discusses the basics of
conditional statements and presents ample illustrations to make things clear.
These conditional statements are used everywhere; from a basic program to
decision support systems and expert systems. The reader is required to go
through the points to remember, and implement the problems given in the
exercise for better understanding of this. One must also understand that the
conditional statements are the first step towards programming. However
understanding conditional statements, though essential, is just the beginning.
Your journey of becoming a programmer has just started.
GLOSSARY
1. The syntax of “if” is as follows.
2. The “if else” construct is written as follows.
if <test condition>
else:
<block if the test condition is not true>
...
3. The syntax of the “if else ladder”
elif <test 2>:
<second block>
elif <test 3>:
<third block>
else:
POINTS TO REMEMBER
The 'if' statement implements conditional branching.
The test condition is a Boolean expression which results in a true or a
false.
The block of 'if' executes if the test condition it true.
The else part executes if the test condition is false.
Multiple branches can be implemented using the if-elif ladder.
Any number of if-elif can be nested.
A ternary if can be implemented in Python.
Logical operators can be used in implementing conditional statements.
EXERCISES
1. What will be the output of the following?

if 28:
print('Hi')
else:
print('Bye')
(a) Hi
(b) Bye
(d) The above snippet will not compile
2. a = 5
b = 7
c = 9
if a>b:
if b>c:
print(b)
else:
print(c)
else:
if b>c:
print(c)
else:
print(b)
(a) 7
(b) 9
(c) 34
(d) None of the following
3. a = 34
b = 7
c = 9
if a>b:
if b>c:
print(b)
else:
print(c)
else:
if b>c:
print(c)
else:
print(b)
(a) 7
(b) 9
(d) The code will not compile
4. a = int(input('First number\t:'))
b = int(input('Second number\t'))
c = int(input('Third number\t:'))
if ((a>b) & (a>c)):
print(a)
elif ((b>a) &(b>c)):
print(b)
else:
print(c)
(a) The greatest of the three numbers entered by the user
(b) The smallest of the three numbers entered by the user
(c) None
5. n = int(input('Enter a three digit number\t:'))

if (n%10)==(n/100):
print('Hi')
else:
print('Bye')
# The three digit number entered by the user is 453
(a) Hi
(b) Bye
6. In the above question, if the number entered is 545, what would the answer
be?
(a) Hi
(b) Bye
7. hb1 = ['Programming in C#','Oxford University Press', 2014]

hb2 = ['Algorithms', 'Oxford University Press', 2015]
if hb1[1]==hb2[1]:
print('Same')
else:
print('Different')
(a) same
(b) Different
(c) No output
(d) The code would not compile
8. hb1 = ['Programming in C#','Oxford University Press', 2014]
if (hb1[0][3]==hb2[0][3]):
print('Same')
else:
print('Different')
(a) Same
(b) Different
(c) No output
9. In the snippet given in question 8, the following changes are made. What
will the output be?
hb1 = ['Programming in C#','Oxford University Press', 2014]
if (str(hb1[0][3])==str(hb2[0][3])):
print('Same')
else:
print('Different')
(a) Same
(b) Different
(c) No output
10. Finally, the code in question 8 is changed to the following. What will the
output be?
hb1 = ['Programming in C#','Oxford University Press', 2014]
if (char(hb1[0][3])==char(hb2[0][3])):
print('Same')
else:
print('Different')
(a) Same
(b) Different
(c) No output
PROGRAMMING EXERCISE
1. Ask the user to enter a number and find the number obtained by reversing
the order of the digits.
2. Ask the user to enter a four digit number and check whether the sum of the
first and the last digits is same as the sum of the second and the third digits.
3. In the above question if the answer is true then obtain a number in which the
second and the third digit are one more than that in the given number.
Example: Number 5342, sum of the first and the last digit = 7 that of the
second and the third digit = 7. New number: 5452
4. Ask the user to enter the concentration of hydrogen ions in a given solution
(C) and find the PH of the solution using the following formula.
PH = log10 C
5. If the PH is <7 then the solution is deemed acidic, else it is deemed as basic.
Find if the given solution is acidic.
6. In the above question find whether the solution is neutral. (A solution is

neutral if the PH is 7)
7. The centripetal force acting on a body (mass m), moving with a velocity v,
in a circle of radius r, is given by the formula mv2/r. The gravitational force
on the body is given by the formula (GmM)/R2, where m and M are the
masses of the body and earth and R is the radius of the earth. Ask the user to
enter the requisite data and find whether the two forces are equal or not.
8. Ask the user to enter his salary and calculate the TADA, which is 10% of the
salary; the HRA, which is 20% of the salary and the gross income, which is
the sum total of the salary, TADA and the HRA.
9. In the above question find whether the net salary is greater than $300,000.
10. Use the Tax Tables of the current year to find the tax on the above income
(question number 8), assuming that the savings are $100,000.
11. Find whether a number entered by the user is divisible by 3 and 13.
12. Find whether the number entered by the user is a perfect square.
13. Ask the user to enter a string and find the alphanumeric characters from the
string.
14. In the above question find the digits in the strings.
15. In question 13, find all the components of the string which are not digits or
alphabets.
CHAPTER
4
LOOPING
• Understand the importance and use of loops
• Appreciate the importance of the while and for
• Use range
• Process list of lists
• Understand nesting of loops and design patterns
4.1 INTRODUCTION
When we were young, we were taught tables of numbers. The table of a number
had a pattern. Writing a table in an examination required writing, say, “n×”
followed by “i” (i varying from 1 to n) and then the result of calculations (that is
n × 1, n × 2 and so on). Many such situations require us to repeat a given task
many times. This repetition can be used to calculate the value of a function, to
print a pattern or to simply repeat something. This chapter discusses loops and
iterations, which are an integral part of procedural programming. Looping means
repeating a set of statements until a condition is true. The number of times this
set is repeated depends on the test condition. Also, what is to be repeated needs
to be chalked out with due deliberation. In general, repeating a block requires the
following (Figure 4.1).
FIGURE 4.1 Looping
Python provides two types of loops: for and while (Figure 4.2).
FIGURE 4.2 Loops in Python
While loop is one of the most general constructs in any programming language.
If you come from a “C” background, you must be equipped with the above
construct. While loop retains most of its features in Python as well, however,
there are notable differences too.
The while loop repeats a block, identified by indentation, until the test condition
remains true. As we will see in the following discussion, one can come out of the
loop using break and continue. Also, to decide if the loop repeats as per the test
condition after which the else condition executes. This is an additional feature
in Python.
The use of for in Python is a bit different to “C”-like languages. The for
construct in Python is generally used for lists, tuples, strings, etc. The chapter
presents range, which will help the programmer to select a value from a given
range. The reader is advised to go through the discussion of lists and tuples
presented in Chapter 2 of this book before starting with the for loop.
The chapter has been organized as follows. Section 4.2 of this chapter presents
the basics of the while loop. Section 4.3 uses looping to create patterns. Section
4.4 introduces the concept of nesting and presents the processing of lists and
tuples using for loops. The last section concludes the chapter.
4.2 WHILE
In Python, the while loop is the most commonly used construct for repeating a
task over and over again. The task is repeated until the test condition remains
true, after which the loop ends and if the exit occurs without a break, then the
else part of the construct executes. The syntax of the loop is as follows:
Syntax
while test:
...
...
else:
...
It may be stated here that the body of the loop is determined by indentation. This
is the reason why you must be extremely careful with indentation. Also, the else
part is an addition in Python when compared to “C”-like languages. In order to
understand the concept, let us go through the following illustrations.
Illustration 4.1: Ask the user to enter a number and calculate its factorial.
Solution: The factorial of a number n is defined as follows.
factorial = 1 × 2 × 3 × … × n
That is the factorial of a number, n, is the product of n terms starting from 1. To

calculate the factorial of a given number, first of all the user is asked to input a
number. The number is then converted into an integer. This is followed by the
initialization of “factorial” by 1. Then a while loop successively multiplies i
to 'factorial' (note that after each iteration the value of i increases by 1). The
following program calculates the factorial of a number entered by the user.
Program
n = input('Enter number whose factorial is required')#ask
user to enter number
m = int(n)#convert the input to an integer
factorial = 1#initialize
i=1# counter
while i<=m:
factorial =factorial*i
i=i+1
print('\factorial of '+str(m)+' is '+str(factorial))
Output
>>>
RUN C:/Users/ACER
ASPIRE/AppData/Local/Programs/Python/Python35-
32/Tools/scripts/factorial.py
Enter number whose factorial is required6
Factorial of 6 is 720
Illustration 4.2: Ask the user to enter two numbers “a” and “b” and calculate
“a” to the power of “b.”
Solution: “a” raised to the power of “b” can be defined as follows.
power = a × a × a × .. × a (b times)
That is, the power of a number “a” raised to “b” is the product of the number
“a,” “b” times. To calculate the power, first of all the user is asked to input two
numbers. The numbers are then converted into integers. This is followed by the
initialization of 'power' by 1. Then a while loop successively multiplies 'a' to
'power' (note that after each iteration the value of i increases by 1). The
following program implements the above logic.
Program
>>>
a = int(input('Enter the first number'))
b = int(input('Enter the second number'))
power=1
i = 1
while i < = b:
power = power*a
i=i+1
else:
print(str(a)+' to the power of '+str(b)+' is
'+str(power))
Output
>>>
RUN C:/Users/ACER
32/Tools/scripts/power.py
Enter the first number4
Enter the second number5
4 to the power of 5 is 1024
>>>
Illustration 4.3: The arithmetic progression is obtained by adding the common

difference “d” to the first term “a,” successively. The ith term of the arithmetic
progression is given by the following formula:
T (i) = a + (i – 1) × d
Ask the user to enter the value of “a,” “d,” and “n” (the number of terms), and
find all the terms of the AP. Also, find the sum of all the terms.
Solution: The following program asks the user to enter the values of “a,” “d,”
and “n.” Note that the input is converted into integers. Also, since all the terms
are to be calculated, this evaluation is done inside a loop. The 'sum' is initialized
to 0 and the terms are added to 'sum' in each iteration.
Program
>>>
a = int(input('Enter the first term of the Arithmetic
Progression\t:'))
d = int(input('Enter the common
difference\t:')) n = int(input('Enter the number of
terms\t:')) i = 1
sum = 0#initialize
while i<=n:
term = a +(i-1)*d
print('The '+str(i)+'th term is '+str(term))
sum = sum + term
i=i+1
else:
print('The sum of '+str(n)+' terms is\t:'+str(sum))
Output
RUN C:/Users/ACER
32/Tools/scripts/AP.py
Enter the first term of the Arithmetic Progression :5
Enter the common difference :6
Enter the number of terms :7
The 1th term is 5
The 2th term is 11
The 3th term is 17
The 4th term is 23
The 5th term is 29
The 6th term is 35
The 7th term is 41
The sum of 7 terms is :161
Illustration 4.4: The geometric progression is obtained by multiplying the

common ratio 'r' to the first term 'a', successively. The ith term of the
progression is given by the following formula. T(i) = a × ri – 1
Ask the user to enter the value of 'a', 'r', and 'n'(the number of terms), and
find all the terms of the GP. Also, find the sum of all the terms.
Solution: The following program asks the user to enter the values of 'a', 'r',
and 'n'. Since all the terms are to be calculated, this evaluation is done inside a
loop. The 'sum' is initialized to 0 and the terms are added to 'sum' in each
iteration.
Program
>>>
a = int(input('Enter the first term of the Geometric
Progression\t:'))
r = int(input('Enter the common ratio\t:'))
n = int(input('Enter the number of terms\t:'))
i = 1
sum = 0#initialize
while i<=n:
term = a * (r**(i-1))
print('The '+str(i)+'th term is '+str(term))
sum = sum + term
i=i+1
else:
print('The sum of '+str(n)+' terms is\t:'+str(sum))
Output
>>>
RUN C:/Users/ACER
ASPIRE/AppData/Local/Programs/Python/Python 35-
32/Tools/scripts/GP.py
Enter the first term of the Arithmetic Progression :5
Enter the common ratio 3
Enter the number of terms 5
The 1th term is 5
The 2th term is 15
The 3th term is 45
The 4th term is 135
The 5th term is 405
The sum of 5 terms is 605
>>>
4.3 PATTERNS
Have you ever wondered why quizzes and riddles form an integral part of any
intelligence test? The following incident will help the reader to understand the
importance of patterns. During World War II, the British were striving hard to
break Enigma, the machine used by the Germans for encrypting their messages.
The army somehow recruited Alan Turing, who was never in his lifetime
recognized, for the above task. He wanted a team to help him, for which he
conducted an exam. Many of you would be amazed to know what he asked in
that exam which would determine the destiny of a country! He asked the
candidates to solve the given puzzles in a given time. This incident underlines
the importance of comprehending patterns. What happened thereafter is history.
Decoding patterns and solving puzzles helps to judge the intellect of a person.
This is much more important than learning a formula. This section presents the
designing of patterns using loops to help the reader understand the concept of
nesting. Moreover, this book also intends to inculcate the problem solving
approach in the reader. Therefore this section becomes all the more important.
The following illustrations show how to assign values to the counters of the
inner and the outer loops to carry out the given task. The patterns, as such, may
not be very useful. However, doing the following program would help the reader
to comprehend the concept of nesting. The methodology of making a pattern has
been explained in each of the following programs.
Illustration 4.5: Write a program to generate the following pattern in Python.
The number of rows would be entered by the user.
Solution: The number of rows n, will determine the value of the counter (from 0
to n). The value of i denotes the row number in the following program. In each
row, the number of stars is equal to the row number. The values of j, in each
iteration, denotes the number of stars in each row. This loop is therefore nested.
Also, note that after the inner loop ends a new line is printed using the print()
function.
Program
Output
RUN C:/Users/ACER
32/Tools/scripts/loop2.py
Enter the number of rows 5
*
* *
* * *
* * * *
1
2 2
3 3 3
4 4 4 4
Solution: The number of rows will determine the value of the counter i, (from 0
to n). The value of i denotes the row number in the following program. In each
row, the number of elements is equal to the row number. The values of j in each
iteration denote the number of elements in each row. This loop is therefore
nested. The element printed is the value of i+1. Also, note that after the inner
loop ends a new line is printed using the print() function.
Program
Output
RUN C:/Users/ACER
Enter the number of rows5
1
2 2
3 3 3
4 4 4 4
5 5 5 5 5
2
2 3
2 3 4
2 3 4 5
Solution: The number of rows, entered by the user, will determine the value of i
(from 0 to n). The value of i denotes the row number in the following program.
In each row, the number of elements is equal to the row number. The values of j
in each iteration denote the number of elements in each row. This loop is
therefore nested. The element printed is the value of j+1. Also note that after the
inner loop ends a new line is printed using the print() function.
Program
Output
>>>
RUN C:/Users/ACER
2
2 3
2 3 4
2 3 4 5
2 3 4 5 6
1
2 3
4 5 6
7 8 9 10
Solution: The value of i denotes the row number in the following program. In
each row, the number of elements is equal to the row number. The values of i in
each iteration will denote the number of elements in each row. This loop is
therefore nested. The element printed is the value of k, which starts from 1 and
incrementally increases in each iteration. Also note that after the inner loop ends
a new line is printed using the print() function.
Program
Output
>>>
RUN C:\Users\ACER ASPIRE\AppData\Local\Programs\Python\
Python35-32\Tools\scripts\loop1.py
1
2 3
4 5 6
7 8 9 10
11 12 13 14 15
16 17 18 19 20 21
22 23 24 25 26 27 28

*
***
*****
*******
*********
Solution: The value of i denotes the row number in the following program. In
each row, the number of stars is equal to the row number. The values of k in each
iteration denote the number of stars in each row, which ranges from 0 to (2*i
+1). This loop is therefore nested. The leading spaces are governed by the value
of j, which ranges from 0 to (m-i-1). This is because if the value of i is 0, the
number of spaces should be 4 (if the value of n is 5). In case the value of i is 1,
the number of spaces should be 3 and so on. Also note that after the inner loop
ends a new line is printed using the print() function.
Program
Output
RUN C:/Users/ACER
Enter the number of rows 6
*
***
*****
*******
*********
4.4 NESTING AND APPLICATIONS OF LOOPS IN LISTS

Nested loops can be used to generate matrices. In order to do this, the outer loop
is designed to govern the rows and the inner loop to govern each element of a
particular row. The following illustration shows the generation of a matrix
having ith element given by the following formula:
ai, j = 5 × (i + j)2
Note that in the following illustration, two loops have been used. The outer loop
runs n times where n is the number of rows, and the inner loop runs m times
where m is the number of columns. The number of columns can be perceived as
the number of elements in each row.
The inner loop has one statement, which calculates the element. At the end of
each iteration (of the outer loop) a new line is printed using the print()
function.
Illustration 4.10: Generate a n × m, matrix, wherein each element (aij), is given

by
ai, j = 5 × (i + j)2
Solution: The concept has been explained in the above discussion. There will be
two loops; the outer loop for the number of rows and the inner loop for the
number of columns.
Program
Output
RUN C:/Users/ACER
32/Tools/scripts/matrixgeneartion.py
Enter the number of columns3
0 5 20
5 20 45
20 45 80
>>>
It may be noted that in the following chapters, this nesting is used to deal with
most of the operations of matrices. As a matter of fact addition and subtraction
of two matrices requires two levels of nesting, whereas multiplication of two
matrices requires three levels of nesting.
Illustration 4.11: Handling list of lists: Note that in the following program the
first list’s second element is itself a list. Its first element can be accessed by
writing hb[0][1] and the first letter of the first element of the nested list would
be hb[0][1][0].
Program
Output
The following two illustrations handle the list of lists using nested loops. Kindly
note the output and the corresponding mappings.
Illustration 4.12: Handling list of lists using loops: The elements of nested
lists can also be dealt with using nested loops as shown in this illustration.
Program
Output
RUN C:/Users/ACER
32/Tools/scripts/listfor.py
[['Programming in C#', ['Oxford University Press', 2015]],
['SE is everything', ['Obscure Publishers', 2015]]]
0 0 Programming in C#
0 1 ['Oxford University Press', 2015]
1 0 SE is everything
1 1 ['Obscure Publishers', 2015]
>>>
Illustration 4.13: Another illustration of the use of loops in processing nested

lists. The user is expected to observe the output and infer what happened.
Program
Output
RUN C:/Users/ACER
32/Tools/scripts/listfor.py
[['Programming in C#', ['Oxford University Press', 2015]],
['SE is everything', ['Obscure Publishers', 2015]]]
0 0 0 P
0 0 1 r
0 0 2 o
0 0 3 g
0 0 4 r
0 0 5 a
0 0 6 m
0 0 7 m
0 0 8 i
0 0 9 n
0 0 10 g
0 0 11
0 0 12 i
0 0 13 n
0 0 14
0 0 15 C
0 0 16 #
0 1 0 Oxford University Press
0 1 1 2015
1 0 0 S
1 0 1 E
1 0 2
1 0 3 i
1 0 4 s
1 0 5
1 0 6 e
1 0 7 v
1 0 8 e
1 0 9 r
1 0 10 y
1 0 11 t
1 0 12 h
1 0 13 i
1 0 14 n
1 0 15 g
1 1 0 Obscure Publishers
1 1 1 2015
4.5 CONCLUSION
Repeating a task is an immensely important job. This is needed in a whole
variety of situations to accomplish different tasks. This chapter introduces the
two most important looping constructs in Python. The chapter demonstrated the
use of these looping constructs by showing simple examples. Having a loop
within a loop is called nesting. The nesting of loops has been explained using
patterns and lists of lists. Chapter 6 revisits one of the constructs and compares
the use of iterators and generators. The reader is expected to solve the problems
given at the end of the chapter for better understanding. However, Python
provides us with other constructs which greatly simplify program writing. At the
moment try various permutations and combinations, observe the outputs and
learn.
GLOSSARY
1. Looping means repeating a task a certain number of times.
2. Syntax of for loop

for i in range(n):
...
...
OR
for i in range(n, m):
...
...
OR
for i in (_, _,...)
...
...
...
3. Syntax of while loop

while <test condition>:
...
POINTS TO REMEMBER
In order to repeat a set of statements a certain number of times looping is
used.
Looping in Python can be implemented using while and for.
'while' is the most common looping construct in Python.
The statements in the while block executes until the test condition remains
true.
The else part executes if the loop ends without a break.
'for' can be used for all the purposes for which a 'while' is used.
'for' is generally used for processing lists, tuples, matrices, etc.
range (n) means values from 0 to (n – 1).
range (m, n) means all the values from m to (n – 1).
A loop can be nested in a loop.
There can be any number of nestings, although this is undesirable.
EXERCISES

a=8
i=1
while a:
print(a)
i=i+1
a=a-i
print(i)
(a) 8, 6, 3
(b) 8, 6, 3, –1
(c) 8, 6, 3, –1, ...
2. a=8
i=1
while a:
print(a)
i=i+1
a=a/2
print(i)
(a) 8, 4, 2, 1
(b) 8, 4, 2, 1, 0
(c) 8, 4, 2, 1, 0.5
(d) Infinite loop
3. How many times will the following loop execute?

n = int(input('Enter number'))
i = n
while (i>0):
print(n)
i=i+1
n = int(n/2)
print(i)
#The value of n entered by the user is 10
(a) 4
(b) 5
(c) Infinite
4. Which loop can be used when the number of iterations is not known?
(a) while
(b) for
(c) both
5. How many levels of nesting are possible in for?

(a) 2
(b) 3
(c) Both
6. n = int(input('Enter number'))
for i in (0,7):
print('i is '+str(i))
i = i+1;
else:
print('bye')
How many values would be printed?
(a) 2
(b) 3
(c) 6
7. n = int(input('Enter number'))
for i in range(n, 1, -1):
for j in range(i):
print(i, j)
#value entered by the user is 5
(a) (5, 0), (5, 1), ...(2, 1)
(b) (5, 1), (5,2),...(2, 0)
(c) (0,1), (0,2), ...(5, 2)
8. In order to print the elements of a given matrix which of the following is

essential?
(a) Nested loops
(b) Single loop
(c) if-else
9. What is meant by range (5)?

(a) Integers from 0 to 4
(b) Integers from 0 to 5
(c) Integers from 1 to 4
(d) Integers from 1 to 5
10. What is meant by range (3, 8)?

(a) 3, 4, 5, 6, 7, 8
(b) 3, 4, 5, 6, 7
(c) 1, 2, 4, 5, 6, 7, 8
(d) 8, 8, 8
PROGRAMMING
1. Ask the user to enter a number and find whether it is a prime number.
2. Ask the user to enter a number and find all its factors.
3. Find whether the number entered by the user is a perfect square.

Example: If number = 30, then factors are 2, 3, and 5
4. Ask the user to enter two numbers and find the lowest common multiple.
Example: If numbers are 30 and 20, then LCM is 60, as both 20 and 30 are
factors of 60
5. Ask the user to enter two numbers and find the highest common factor.
Example: If numbers are 30 and 20, the HCF is 10
6. Find the mean of numbers entered by the user.
7. Find the variance and standard deviation of the numbers entered by the user.
8. Ask the user to enter the values of a and b and find aba.
9. Find the common factor of n numbers entered by a user.

10. Ask the user to enter three numbers and find all possible permutations of
these numbers.
11. In the above question, what happens if we have four numbers in place of
three?
12. Can the above logic be extended for n numbers?
13. Ask the user to enter n numbers and find the minimum of the numbers
without using arrays.
14. Ask the user to enter n numbers and find the maximum of the numbers
without using arrays.
15. Create a list of authors in which the record of each author is itself a list
consisting of the name of the book, publisher, year of publication, ISSN, and
the city. Now process the list using for loop.
CHAPTER
5
FUNCTIONS
• Appreciate the importance of modular programming
• Define and classify functions
• Understand the concept of the scope of a variable
• Understand and use recursion
5.1 INTRODUCTION
If you have to perform a bigger task, then it is advisable to divide it into smaller,
more manageable tasks. This division has many advantages, which are discussed
in the following sections. The units of programs, which can be called on as its
basis, take some input, process it and may generate some output which i referred
to as functions.
Functions are units which perform a particular task, take some input, and
which may give some output.
This concept is the soul of procedural programming. The readers familiar with C
(or for that matter C++, Java, C#, etc.) will be familiar with the idea and use of
functions. However, a brief discussion on the futures and advantages of
functions follows in the next section.
This chapter introduces the concept of functions. The chapter has been organized
as follows. The next section briefly explains the features of a function; the third
section explains the basic terminology, and the following section discusses the
definition and use of a function. The fifth section presents a brief discussion on
the scope of a variable. The sixth section presents recursion and the last section
of the chapter concludes.
5.2 FEATURES OF A FUNCTION
As discussed earlier, functions form the basis of procedural programming. One
of the most obvious advantages is the division of a program into smaller parts.
This section briefly discusses the advantages of dividing the program into
functions.
5.2.1 Modular Programming

If a program is divided into small parts in such a way that different parts each
perform some specific task, then each part can be called as per the requirement.
5.2.2 Reusability of Code

A function can be called many times. This spares the programmer from the
horror of rewriting the same code again, which in turn can reduce the length of
the program.
5.2.3 Manageability
Dividing a bigger task into smaller functions makes the program manageable. It
becomes easy to locate bugs and therefore make the program reliable. It also
becomes easy to carry out local optimization in a function. To summarize,
manageability leads to the following:
5.2.3.1 Easy Debugging
In order to understand why creating functions will make debugging easy, let us
consider White Box Testing. This type of testing, which uses code for testing,
requires elicitation of paths and crafting test cases catering to them. In this case
it becomes easy to effectively analyze smaller functions rather than the whole
task.
5.2.3.2 Efficient
It is essential to make code efficient both in terms of time and memory. As a

matter of fact, even in “C’s” compiler most of the code optimization is attributed
to the developer rather than the compiler.
The above factors point to the fact that dividing the task into functions is good
practice. It may be noted here that even object-oriented programming, described
in Section 2 of this book, relies on functions for implementing the behavior of a
class.
5.3 BASIC TERMINOLOGY

The importance of functions in procedural programming has already been
discussed in the previous section. This section briefly introduces the terminology
of functions and presents the syntax which will form the foundation stone of the
discussion that follows.
5.3.1 Name of a Function

A function can have any legal literal name. For example, sum1 is a valid
function name as it satisfies all the requisite constraints. It may be stated here
that in a class we can have more than one function with the same name and
different parameters. This is referred to as overloading. The concept has been
discussed in Section 2 of this book.
5.3.2 Arguments
The arguments of a function denote the input given to a function. A function can
have any number of arguments. As a matter of fact, it is possible that a function
may not have any argument.
5.3.3 Return Value

A function may or may not return a value. The beauty of Python lies in not
specifying the return type and therefore using the same functions for various data
types.
In Python a function can be made in the command prompt. This implies that
unlike C (or for that matter C++, Java, or C#) a function need not be a part of a
program. Moreover, the return type as described in this section need not be
mentioned. This inculcates flexibility in the procedures.
5.4 DEFINITION AND INVOCATION

This section discusses how to define a function and call a function that has been
defined. The definition of a function depicts the behavior of the function. The
task to be performed by the function is contained in the definition. In the
discussion that follows, the components of a definition have been explained in
detail.
The invocation of a function refers to calling a function. As explained in Section

5.6, a function can also be called within itself. This is referred to as recursion. It
may also be noted that a function is defined only once. However, it can be called
any number of times.
FIGURE 5.1 Example of a function
The definition of a function contains the following:
Name of a function: The name of a function is any valid identifier. It should be

noted though that the name of a function should be meaningful and, if possible,
convey the task to be performed by the function.
Parameter: The list of parameters (separated by commas) is given in the

parentheses following the name of the function. The parameters are basically the
input to the function. A function may have any parameters.
Body of the function: The body of the function contains the code that
implements the task to be performed by the function.
Figure 5.1 shows the name of the function (fun), the list of parameters in the
parentheses following the name of the function (in this case there are no
parameters) and the body of the function.
It may also be noted that the closing parentheses containing the parameters is
followed by a colon. The body of a function starts with a proper indentation.
The invocation of a function can be at any place after the definition. However,
exceptions to this premise are found in the case of recursion.
The syntax of a function is depicted in Figure 5.2.
Syntax
FIGURE 5.2 Syntax of a function
5.4.1 Working
Consider a function which multiplies two numbers passed as parameters.
def product(num1, num2):
prod= num1*num2
print('The product of the numbers is \t:'+str(prod))
The name of this function is product. It takes two arguments as input (num1 and
num2), calculates the product and displays the results.
The function can be invoked as follows:

num1=int(input('Enter the first number\t:'))
num2=int(input('Enter the second number\t:'))
print('Calling the function...')
product(num1, num2)
print('Back to the calling function');
Here calling product shifts the control to the function, inside which the product
is calculated and the result is displayed. The control then comes back to the
calling function (Figure 5.3).
FIGURE 5.3 Calling a function
A function can be called any number of times. The following example shows a
function which does not take any input and does not return anything. The
function called just prints the lines of Ecclesiastes. The following listing shows
the function, and the output of the program follows.
Illustration 5.1: Basic Function
Listing
def Ecclesiastes_3():
print('To everything there is a season\nA time for every
purpose under Heaven')
print('A time to be born\nand a time to die\nA time to
plant\nand a time to reap')
print('A time to kill\nand a time to heal\nA time to
break down\nand a time to build up')
print('A time to cast away stones\nand a time to gather
stones\nA time to embrace\nand a time to refrain')
print('A time to gain\nand a time to lose\nA time to
keep\nand a time to cast away')
print('A time of love\nand a time of hate\nA time of war\
nand a time of peace')
print('Calling function\n')
Ecclesiastes_3()
print('Calling function again\n')
Ecclesiastes_3()
>>>
Output
Calling function
To everything there is a season

A time for every purpose under Heaven
A time to be born
and a time to die
A time to plant
and a time to reap
A time to kill
and a time to heal
A time to break down
and a time to build up
A time to cast away stones
and a time to gather stones
A time to embrace
and a time to refrain
A time to gain
and a time to lose
A time to keep
and a time to cast away
A time of love
and a time of hate
A time of war
and a time of peace
5.5 TYPES OF FUNCTION

Based on the parameters and the return type, functions can be divided into the
following categories. The first type of function does not take any parameter or
return anything. The program given in Illustration 5.1 shows one such function.
The second type of function takes parameters but does not return anything.
Illustration 5.2 is an example of a function. The third type of function takes
parameters and returns an output. The example that follows adds two numbers
using functions. The task has been accomplished in three different ways – in the
first function (sum1) the input is taken inside the function and the result is
displayed in a print statement, which is also present inside the function.
The second function takes the two numbers as input (via parameters), adds them
and prints the result inside the function itself. The third function (sum3) takes
two parameters and returns the sum.
Illustration 5.2: Write a program to add two numbers using functions. Craft
three functions, one of which does not take any parameters and does not return
anything. The second function should take parameters and not return anything.
The third function should take two numbers as parameters and should return the
sum.
Program
>>>
def sum1():
sum= num1+num2
print('The sum of the numbers is \t:'+str(sum))
def sum2(num1, num2):

sum= num1+num2
print('The sum of the numbers is \t:'+str(sum))

sum= num1+num2
return(sum)
print('Calling the first function...')

sum1()
print('Calling the second function...')
sum2(num1, num2)
print('Calling the third function...')
result=sum3(num1, num2)
print(result)
>>>
Output
RUN C:/Users/ACER
32/Tools/scripts/sum_of_numbers.py
Calling the first function...
The sum of the numbers is 7
Calling the second function...
The sum of the numbers is 3
Calling the third function...
3
>>>
5.5.1 Advantage of Arguments

In Python, unlike “C” while defining a function the types of arguments are not
specified. This has the advantage of giving different types of arguments in the
same function. For example in the function that follows, the first invocation
passes an integer value in the function. The function multiplies the number by
two. In the case of the second invocation the multiply operator repeats the string,
passed as a parameter, n number of times.
Illustration 5.3: Types of functions.
Listing 1

return (num1+num2)
Output
>>> sum1(3,2)
5
>>> sum1('hi', 'there')
'hithere'
>>>
Listing 2
>>>
return (num1+num2)
print('Calling function with integer arguments\t: Result:
'+str(sum1(2,3)))
print('Calling the function with string arguments\t: Result:
'+sum1('this',' world'))
Output
Calling function with integer arguments : Result: 5
Calling the function with string arguments : Result: this
world
>>>
5.6 IMPLEMENTING SEARCH

This section demonstrates one of the most important uses of the topics studied so
far: Searching. In the search problem if the element is present in a given list
then its position should be printed, otherwise a message “Not Found” should be
displayed. There are two major strategies used to accomplish the task. They are
linear search and binary search. In linear search, the elements are iterated one by
one. If the required element is found, the position of the element is printed. The
absence of an element can be judged using a flag. The algorithm has been
implemented in Illustration 5.4.
Illustration 5.4: Write a program to implement a linear search.
Solution:
Output
Position 5
Not found
>>>
The above search strategy works well. However, there is another strategy of
search called binary search. In binary search, the input list must be sorted. The
algorithm checks whether the item to be searched is present at the first position,
at the last position or at the middle position. If the requisite element is not
present at any of these positions and it is less than the middle element, then the
left part of the list becomes the input of the procedure; else the right part of the
element becomes the input to the procedure. The reader is advised to implement
binary search.
The complexity of the binary search is O(logn).
5.7 SCOPE
The scope of a variable in Python is the part of the program wherein its value is
legal or valid. It may be seated here that although Python allows a global
variable, the value of a local variable must be assigned before being referenced.
Illustration 5.5 exemplifies this concept. The illustration has three listings. In the
first listing the value of 'a' has been assigned outside the function as well as
inside the function. This leads to a problem as a variable cannot be referenced
before being assigned.
In the second case this contention is resolved. Finally, the last listing shows that
global variables are very much allowed in Python for some strange reason. As an
active programmer, I firmly believe that should not have been allowed and there
are multiple reasons for not allowing global variables in a programming
language.
Illustration 5.5: Listings to scope of a variable.
Listing 1
# Note that a = 1does not hold when function is called
a = 1
def fun1():
print(a)
a=7
print(a)
def fun2():
print(a)
a=3
print(a)
fun1()
fun2()
>>>
Output
============ RUN C:/Python/Functions/scope.py ===========
File "C:/Python/Functions/scope.py", line 12, in
<module> fun1()
File "C:/Python/Functions/scope.py", line 3, in fun1
print(a)
UnboundLocalError: local variable 'a' referenced before
assignment
>>>
Listing 2
a = 1
def fun1():
a=1
print(a)
a=7
print(a)
def fun2():
a=1
print(a)
a=3
print(a)
fun1()
fun2()
Output
>>>
=========== RUN C:/Python/Functions/scope.py ===========
1
7
1
3
>>>
Also, note that had “a” been not assigned in the functions, the global value
would have sufficed.
Listing 3
Output
>>>
=========== RUN C:/Python/Functions/scope.py ============
1
1
>>>
5.8 RECURSION
At times a function needs to be called within itself. Calling a function in itself is
referred to as recursion. The concept is used to accomplish many tasks easily and
intuitively. For example, consider the following series:
1, 1, 2, 3, 5, 8, 13, ...
Note that each term is the sum of the previous two terms; the first and the second
term being 1 and 1 respectively. This series is referred to as the Fibonacci series.
The series is due to a famous rabbit problem which has been described as
follows in mathematics.
5.8.1 Rabbit Problem

A pair of rabbits does not breed for the first two months together, after which
they generate a pair of rabbits each month. This way there is a single pair of
rabbit for the first two months, after which the growth follows a fascinating
sequence (Table 5.1).
Table 5.1 Fibonacci series
Month Pair of Rabbits Number of

Pairs
1 R0 1
2 R0 1
3 R0 R01 2
4 R0 R01, R02 3
5 R0 R01 ( R010), R02, R03 5
6 R0 R01 ( R010, R011), R02 ( R020), R03, 8
R04
Note that in the above series, each term is the sum of the two preceding terms.
The series can be represented as follows:
Illustration 5.6 depicts the implementation of the Fibonacci series using

recursion.
Illustration 5.6: Ask the user to enter the value of n and find the nth Fibonacci
term.
Solution:
Output
========== RUN C:/Python/Functions/factoorial.py ==========
Enter the number:5
The nth fib term is 5
Note that the calculation of Fibonacci uses the Fibonacci term calculated earlier.
For example, the calculation of the 5th Fibonacci term requires the following
calculations: fib(5) requires fib(4) and fib(3), fib(4) requires fib(3) and fib(2) and
fib(3) requires fib(2) and fib(1) (Figure 5.4).
FIGURE 5.4 Calculation of the 5th Fibonacci term
The next example calculates the factorial of a number using recursion. The
factorial of a number n (positive, integer) is the product of all the integers from 1
to n. That is:
n! = 1 × 2 × 3 × ... × n
Note that since (n – 1)! = 1 × 2 × 3 ... × (n – 1)
Therefore, n! = n × (n – 1)!
Also the factorial of 1 is 1, that is, 1! = 1 which can be used as the base case
while implementing factorial using recursion. The program has been depicted in
Illustration 5.7.
Illustration 5.7: Ask the user to enter the value of n and calculate the factorial
of n using recursion.
Solution:
Output
>>>
========== RUN C:/Python/Functions/factoorial.py ==========
Enter the number 5
Factorial of 5 is 120
>>>
The power of a number raised to the power of another can also be calculated
using recursion. Since that is power (a, b) = a*power (a, b – 1). Also, a1,
that is power (a, 1) = 1. The above logic has been implemented in the
illustration that follows.
Illustration 5.8: Ask the user to enter the values of a and b and calculate a to the
power of b, using recursion.
Program
def power(a, b):
if b==1:
return a
else:
return (a*power(a, b-1))
a = int(input('Enter the first number\t:'))
b = int(input('Enter the second number\t:'))
p = power(a,b)
print(a, ' to the power of ',b,' is ', p)
Output
>>>
========== RUN C:/Python/Functions/power.py ==========
3 to the power of 4 is 81
>>>
>>>
5.8.2 Disadvantages of Using Recursion

In spite of the fact that recursion makes things easy and helps to accomplish
some of the tasks intuitively, there is a flip side. Consider the first illustration.
Though the program calculates the nth Fibonacci term easily, the complexity of
the procedure is too high (O( )) where Φ is gold number. The same task can be
accomplished in linear time using a paradigm called dynamic programming.
Similarly, the recursive procedures in divide and conquer also require a huge
amount of time. In addition to the above problem, there is another flip side.
Recursion requires a lot of memory. Though a portion of the memory is reserved
for stacks, a recursive procedure may eat up all the available memory. However,
recursion is fun so let’s enjoy recursion.
5.9 CONCLUSION
The chapter introduces the concept of functions. The idea of dividing the given
program into various parts is central to manageability. The chapter forms the
foundation stone of the chapters that follow. It may also be stated that function
implements the behavior of a class; therefore before moving to the object-
oriented paradigms, one must be familiar with functions and procedures.
The concept of recursion is also central to the implementations which involve

the ideas of divide and conquer and that of dynamic programming. So, one must
also be equipped with the power of recursion and should be able to use the
concept to solve problems if possible.
The discussion continues in the next chapter where the ideas of iterators,
generators, and comprehensions have been introduced. As a matter of fact, all of
them are functions but with special purposes.
GLOSSARY
Function: Functions accomplish a particular task. They help with making a
program manageable.
Argument: Arguments are the values passed in a function.
Recursion: A function may call itself. This is referred to as recursion.
POINTS TO REMEMBER
A function can have any number of arguments.
A function may return a maximum of one value.
A function may not even return a value.
A function may call itself.
A function can be called any number of times.
A function needs to be called in order to be able to accomplish a particular
task.
EXERCISES
1. Which of the following keywords is used to define a function?

(a) Def
(b) Define
(c) Definition
2. The values passed in a function are called

(a) Arguments
(b) Return values
(c) Yield
3. A recursive function is one that calls

(a) Itself
(b) Other function
(c) The main function
4. Which of the following should be present in a recursive function?

(a) Initial values
(b) Final values
(c) Both
5. Which of the following can be accomplished using recursion?

(a) Binary search
(b) Fibonacci series
(c) Power
6. Which of the following is allowed in a function?

(a) if
(b) while
(c) Calling a function
7. Which types of functions are supported in Python?

(a) Build in
(b) User defined
(c) Both
8. Which of the following is true?

(a) A function helps in dividing a program in small parts
(b) A function can be called any number of times
(c) Both

(a) One can have a function that called any number of functions
(b) Only a limited number of functions can be called from a function
(c) Nested functions are not allowed in Python
(d) Nested functions are allowed only in certain conditions
10. Nested functions incorporates the concept of

(a) Stack
(b) Queue
(c) Linked List
1. Write a function that calculates the mean of numbers entered by the user.
2. Write a function that calculates the mode of numbers entered by the user.
3. Write a function that calculates the median of numbers entered by the user.
4. Write a function that calculates the standard deviation of the numbers

entered by a user.
5. Write a function that finds the maximum of the numbers from a given list.
6. Write a function that finds the minimum of the numbers from a given list.
7. Write a function that finds the second maximum of the numbers from a
given list.
8. Write a function that finds the maximum of three numbers entered by the
user.
9. Write a function that converts the temperature in Celsius to that in

Fahrenheit.
10. Write a function that searches for an element from a given list.
11. Write a function that sorts a given list.
12. White a function that takes two lists as input and returns the merged list.
13. Write a function that finds all the factors of a given number.
14. Write a function that finds common factors of two given numbers.
15. Write a function that returns a number obtained by reversing the order of
digits of a given number.
QUESTIONS BASED ON RECURSION

Use recursion to solve the following problems
1. Find the sum of two given numbers.
2. Find the product of two given numbers.
3. Given two numbers, find the first number to the power of the second.
4. Given two numbers, find the greatest common divisor of the numbers.
5. Given two numbers, find the least common multiples of the numbers.
6. Generate n Fibonacci terms.

In a series the first three terms are 1, 1 and 1; the ith term is obtained using
the following formula
f (i) = 2 × f (i – 1) + 3 × f (i – 2)
Write a function to generate n terms of the sequence.
Find the element in a given sorted list.
Find the maximum from a given list.
Reverse the order of digits of a given number.
THEORY
1. What are the advantages of using functions in a program?
2. What is a function? What are the components of a function?
3. What is the importance of parameter and return type in a function? Can a

function have more than one return value?
4. What is recursion? Which data structure is used internally while

implementing recursion?
5. What are the disadvantages of recursion?
EXTRA QUESTIONS
1. What will be the output of the following program?
(a) 1
(b) 5
(c) 3
(d) Maximum iteration depth reached

(a) 5
(b) 27
(c) Maximum iteration depth reached
(a) 5
(b) 100
(c) 25
(d) Maximum iteration depth reached
(a) 1 2 8 28 100 356 1268 4516 16084

(b) 1 3 5 7 9 11 13 15
(a) 1 2 8 28 100 356 1268 4516 16084

(b) 1 3 5 7 9 11 13 15
(a) I am in main
I am in fun 1
I am in fun 2
I am back in fun 1
I am back in main
(b) Reverse of the above
(d) The program does not execute
7. Conceptually which data structure is implemented in the above program?

(a) Stack
(b) Queue
(c) Graph
(d) Tree
8. Which technique is implemented in the following code?
(a) Linear search

(b) Binary search
(d) The code does not execute
9. What is the complexity of the above?

(a) O (n)
(b) O (n2)
(c) O (log n)
10. Which is better - linear search or binary search?
(a) Linear
(b) Binary
(c) Both are equally good
(d) Depends on the input list.
CHAPTER
6
ITERATIONS, GENERATORS, AND

COMPREHENSIONS
• Understand the use and application of iterators
• Use iterators to produce sequences
• Use generators to generate sequences
• Understand and use list comprehensions
6.1 INTRODUCTION
So far basic data types, operators and control structures have been covered.
These are essential part of any procedural language. Python also provides the
programmers with lists, strings, tuples, dictionary, and files, which makes it a
powerful language. However, one should be able to efficiently access and
manipulate these elements to accomplish the given task. For example if the
formula of the ith element of a list is known, then the whole list can be
generated/accessed in one go. To accomplish this task, for loop comes to our
rescue. However, Python also comes with better options like iterators, which
help us to do the said task easily. One can also define iterable objects in Python.
There is another marvel in Python that facilitates the generation of lists and
sequences, which is generator. This chapter also introduces comprehension.
The chapter has been organized as follows. The second section of this chapter
revisits for. The iterators have been introduced in the third section of this
chapter. The fourth section explains so as to how to define your own iterable
objects. The generators are introduced and explained in section five of this
chapter. The sixth section of this chapter deals with comprehensions, which
makes the task of generating specific lists, etc., easy and the last section
concludes the chapter.
The chapter assumes importance as it forms the foundation stone of many of the
difficult tasks presented in the following chapters. Also, the knowledge of these
will make the day to day tasks easy and spare the programmer from the horror of
writing long codes.
6.2 THE POWER OF “FOR”
A for loop can be used to iterate through a list, tuple, string, or a dictionary. This
section briefly explains how to use the loop for the above iterable objects. Let us
start with the syntax of for.
Syntax
for i in L:
#do something
L is list, string, tuple or dictionary
When one writes " i in L", where L is a list, i becomes the first element of the
list and as the iteration progresses, i becomes the second element, the third
element and so on. These elements can be then independently manipulated. The
concept has been exemplified in Illustration 6.1. The illustration shows the
manipulation of a list using the for loop. In the illustration the given list
contains a set of numbers, some of them positive and some negative. The
negative numbers are appended to a list called N, where the positive numbers are
appended in a list called P.
Illustration 6.1: From a given list, put all the positive numbers in one list and
negative numbers in the other list.
Solution: Create two lists, P and N. Initialize both of them to []. Now check each
number in the list. If the number is positive put it in P and if the number is
negative put it in N.
Program
Output
========== RUN C:/Python/Iterations/for list.py ==========
The list of positive numbers : [1, 2, 5, 7, 3, 7]
The list of negative numbers : [-1, -6]
>>>
A for loop can also be used to manipulate strings. When one writes " i in
str", where str is a string, i becomes the first character of the string and as the
iteration progresses, i becomes the second character, the third character and so
on. These characters can be then independently manipulated. The concept has
been exemplified in Illustration 6.2, in which the vowels and consonants of a
given string are put in two strings.
Illustration 6.2: Ask the user to enter a string and put all the vowels of the
string in one string and the consonants in the other string.
Solution: Create two strings: str1 and str2. Initialize both to "". Now, check
each character in the given string. If it is a vowel, concatenate it with str1
otherwise concatenate it with str2.
Program
Similarly, a for loop can be used to iterate through a tuple and keys of a
dictionary as shown in Illustrations 6.3 and 6.4.
Illustration 6.3: This illustration demonstrates the use of for for iterating
through a tuple.
Solution:
T=(1, 2, 3)
for i in T:
print(i)
print(T)
>>>
Output
========== RUN C:/Python/Iterations/forTuple.py ==========
1
2
3
(1, 2, 3)
>>>
Illustration 6.4: This illustration demonstrates the use of for for iterating
through a dictionary.
Solution:
Dictionary={'Programming in C#': 499, 'Algorithms Analysis
and Design':599}
print(Dictionary)
for i in Dictionary:
print(i)
>>>
Output
========== RUN C:/Python/Iterations/for dic.py ==========
{'Programming in C#': 499, 'Algorithms Analysis and Design':
599}
Programming in C#
Algorithms Analysis and Design
>>>
6.3 ITERATORS
The above tasks (Illustrations 6.1-6.4) can also be accomplished using iterators.
The "iter" function returns the iterator of the object passed as an argument. The
iterator can be used to manipulate lists, strings, tuples, files, and dictionary, in
the same way as a for loop. However, the use of an iterator ensures flexibility
and additional power to a programmer. This will be established in the following
sections.
An iterator can be set on a list using the following:

<name of the iterator> = iter(<name of the List>
The iterator can move to the next element, using the _next_() method. An
iterator, as stated earlier, can iterate through any iterable object including lists,
tuples, strings or a directory. When there are no more elements then a
StopIteration exception is raised.
The following illustration shows the manipulation of a list using iterators. In the
illustration, the given list contains a set of numbers, some of them positive and
some negative. The negative numbers are appended to a list called N, whereas
the positive numbers are appended in a list called P. The same problem was
solved using the for loop in Illustration 6.1.
Illustration 6.5: Using iterators the program puts the positive and negative
numbers of a list into two separate lists and raises an error at the end of the
program.
Solution:
The next example deals with a string. The iterator is set to the first element of
the string and is then set to the second element, third element and so on. If the
character is a vowel it is appended to vow, otherwise it is appended to cons. The
following illustration uses the same problem as that stated in Illustration 6.2.
Illustration 6.6: Write a program that uses iterators to separate the vowels and
consonants of a given string and raises an error at the end of the program.
Solution: The vow and cons strings are initialized to "" and each character of the
given list is checked. If the character is a consonant it is concatenated to cons,
otherwise it is concatenated to vow.
A slightly more complex application of iterators has been shown in the following
illustration.
Illustration 6.7: Write a program to add the corresponding elements of two

given lists and sort the final list.
Solution:
#The program concatenates two lists into one by iterating
over individual elements of the lists using the list
function and then sorts the concatenated list.
l1 = [ 3, 6, 1, 8, 5]
l2 = [ 7, 4, 6, 2, 9]
i1 = iter(l1)
i2 = iter(l2)
l3 = sorted( list(i1) + list(i2) )
print( 'List1 - ', l1, '\nList2 - ', l2, '\nSortedCombn - ',
l3 )
6.4 DEFINING AN ITERABLE OBJECT

One can define a class, in which _init_, _iter_, and _next_ can be defined as
per the requirement. The init function initializes the variables of the class, the
iter defines the mechanism of iterations and the next method implements the
jump to the next item.
Illustration 6.8: Generate the terms of an arithmetic progression using an
iterator by creating an iterable object.
Solution:
Output
=========== RUN C:/Python/Iterations/Class.py ===========
Enter the first term :1
Enter the common differnce :2
<class ' main .yrange'>
1
3
5
>>>
6.5 GENERATORS
Generators are functions that generate the requisite sequences. However, there is
an inherent difference between a normal function and a generator. In a generator,
the values are generated as and when we proceed. So, if one comes back to the
function once a particular value is generated, then instead of starting from the
beginning the function starts from the point where we left off.
The task seems difficult but has an advantage. The concept can help the
programmer to generate lists containing the desired sequences. For example if
one wants to generate a list containing the terms of an arithmetic progression in
which each term is “d” more than the first term, generators come to the rescue.
Similarly, the sequences like geometric progression, Fibonacci series, etc., can
be easily generated using generator.
Python comes with 'yield', which helps to start from the point where we left
off. This is markedly different from 'return' used in normal functions which
does not save the state where we left off. If the function having 'return' is
called again, it starts all over again.
The following illustration exemplifies the use of generators to produce simple

sequences like arithmetic progression, geometric progression, Fibonacci series,
etc.
Illustration 6.9: Write a generator to produce arithmetic progression where in

the first term, the common difference and the number of terms is entered by the
user.
Solution:
Output
========= RUN C:/Python/Iterations/generator 1.py =========
Enter the first term of the arithmetic progression :3
Enter the common difference of the arithmetic progression :5
Enter the number of terms of the arithmetic progression :8
<generator object arithmetic_progression at 0x031C2DE0>
3
8
13
18
23
28
33
38
>>>
Illustration 6.10: Write a generator to produce geometric progression, where in

the first term the common ratio and the number of terms is entered by the user.
Solution:
Output
======== RUN C:/Python/Iterations/generators gp.py ========
Enter the first term of the geometric progression :3
Enter the common ratio of the geometric progression :4
Enter the number o0f terms of the geometric progession :7
3
9
27
81
243
729
2187
>>>
Illustration 6.11: Write a generator to produce a Fibonacci series.
Solution:
In order to understand the concept, let us go through the following illustration.
Illustration 6.12: This illustration demonstrates the effect of yield on the value
of the counter.
Solution: The reader is expected to note the change in the value after and before
yield.
Program
Output
========== RUN C:/Python/Iterations/generator. py ==========
Start
Value of i before yield : 0
0
Value of i after yield : 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
End
>>>
6.6 COMPREHENSIONS
The aim of a programming language should be to make things easy for a
programmer. A task can be performed in many ways but one which requires the
least coding is the most appealing to a coder. Python has many features which
facilitate programming. Comprehensions are one of them. Comprehensions
allow sequences to be built from other sequences. Comprehensions can be used
for lists, dictionary and set comprehension. In the earlier version of Python
(Python 2.0) only list comprehensions were allowed. However, in the newer
versions comprehensions can be used with dictionary and sets also.
The following illustration explains the use of comprehensions to generate lists in

various cases:
The range (n) function generates numbers up to n. The first comprehension
generates the list of numbers which are cubes of all the numbers generated
by the range function.
The second comprehension works in the same way but generates 3 to the
power of x.
The third comprehension generates a list having numbers generated by the
range (n) function, which are multiples of 5.
In the fourth comprehension the comprehension takes the words of the sentence
“Winter is coming” and generates a list containing the word in caps, in running
and the length of the word.
Illustration 6.13: Generate the following lists using comprehensions
x3, i from 0 to 9
3x, i from 2 to 10
All the multiples of 5 from the previous list
The caps, running version and the length of each word in the sentence
“Winter is coming”
Solution:
L1 = [x**3 for x in range(10)]
print(L1)
L2 = [3**x for x in range(2, 10, 1)]
print(L2)
L3 = [x for x in L2 if x%5==0]
print(L3)
String = "Winter is comming".split()
print(String)
String_cases=[[w.upper(), w.lower(), len(w)] for w in
String]
for i in String_cases:
print(i)
list1 = [1, '4', 9, 'a', 0, 4]
square_int = [ x**2 for x in list1 if type(x)==int]
print(square_int)
>>>
Output
>>>
====== RUN C:/Python/Iterations/Comprehensions 1.py ======
[0, 1, 8, 27, 64, 125, 216, 343, 512, 729]
[9, 27, 81, 243, 729, 2187, 6561, 19683]
[]
['Winter', 'is', 'comming']
['WINTER', 'winter', 6]
['IS', 'is', 2]
['COMMING', 'comming', 7]
[1, 81, 0, 16]
>>>
A comprehension contains the input sequence along with the expression that
represents the members. A comprehension may also have an optional predicate
expression.
In order to understand the concept let us consider one more illustration. The list
of temperatures in Celsius is given and the corresponding list containing the
temperatures in Kelvin is to be generated. It may be stated here that the
temperatures in Celsius and Kelvin are related as follows.
Kelvin (T) = Celsius(T)+273.16
Illustration 6.14: Given a list containing temperatures in Celsius, generate a list

containing temperatures in Kelvin.
Solution: The list L_kelvin, is a list where in each element is 273.16 more than
the corresponding element in L_cel. Note that the task has been accomplished in
the definition of the list L_Kelvin itself.
Program
L_Cel = [21.2, 56.6, 89.2, 90,1, 78.1]
L_Kelvin = [x +273.16 for x in L_Cel]
print('The output list')
for i in L_Kelvin:
print(i)
Output
>>>
====== RUN C:/Python/Iterations/comprehension_cel.py ======
The output list
294.36
329.76000000000005
362.36
363.16
274.16
351.26
>>>
Another important application of comprehension is to generate the Cartesian

product of two sets. The cross product of two sets, A and B, is a set containing
tuples of the form (x, y), where x belongs to the set A and y belongs to the set B.
Illustration 6.15 implements the program.
Illustration 6.15: Find the Cartesian product of two given sets.
Solution:
A= ['a', 'b', 'c']
B= [1, 2, 3, 4]
AXB = [(x, y) for x in A for y in B]
for i in AXB:
print(i)
>>>
Output
======== RUN C:/Python/Iterations/cross_product.py ========
('a', 1)
('a', 2)
('a', 3)
('a', 4)
('b', 1)
('b', 2)
('b', 3)
('b', 4)
('c', 1)
('c', 2)
('c', 3)
('c', 4)
>>>
The above program is important because the concept of relations and therefore
functions in mathematics originates from the cross product. As a matter of fact
any subset of A × B is a relation from A to B. There are four types of relations in
mathematics: one to one, one to many, many to one, and many to many. Out of
these relations, one to one and many to one are referred to as functions.
6.7 CONCLUSION
The chapter explains the use of for for iterating over a list, string, tuple, or a
dictionary. It may be stated here that in C or C++, for is generally used for the
same purpose as while. However in Python, for can be used to visit each
element individually. Note that this can also be done in Java or C#. In order to
define an iterable object, _iter and _next need to be defined for the requisite
class. The reader is also expected to take note of the fact that yield and return
perform different tasks in Python. The use of these two has been demonstrated in
the illustrations presented in this chapter. Finally, while defining a list each
element can be crafted as per the need of the question. This can be done with the
help of comprehensions. The chapter, though easy, becomes important in the
light of excessive use of these techniques in machine learning and pattern
recognition tasks which are introduced in the last section of this book.
GLOSSARY
Iterator takes an iterable object and helps to traverse the object.
_next_(): The next function helps in iterating over the value of the iterable
object.
_iter_(): It helps in creating a user defined iterable object.
yield(): It does not return anything.
POINTS TO REMEMBER
for statement can be used for looping over a list, string, tuple, file, and
dictionary
iter takes an object and returns the corresponding iterator
The _next_ gives us the next element
Built in functions, lists, etc., accept iterator as arguments
A generator produces a sequence of results
Yield is used when many values are to be produced from a
function/generator
EXERCISES
1. Which of the following can be an argument in _iter()_?

(a) String
(b) Tuple
(c) List
(d) Dictionary
(e) All of the above
2. The iter takes which type of object?

(a) Iterable
(b) Any object
(c) Comprehension
(d) Generator
3. What is the function of _next()_?

(a) To iterate over the items of an iterable object.
(b) To produce a new iteration
(c) To iterate through a generator
4. Which of the following transfers the control to the calling function?
(a) return
(b) yield
(c) Both
5. Which of the following does not transfer the control to the calling function?
(a) return
(b) yield
(c) Both
6. Which of the following is essentially used in generators?

(a) yield
(b) return
(c) Both

(a) One can use iterators with generators
(b) One can use iterators with list
(c) One can use iterators with comprehensions
8. Which of the following can be iterated using a for loop

(a) String
(b) List
(c) Tuple
9. Which of the following can be iterated using a for loop
(a) String
(b) Comprehension
(c) File
10. Which of the following behaves in the same manner as the combination of
_iter()_ and _next()_?
(a) for
(b) if
(c) Both
THEORY
1. Explain how a for can be used to iterate over an iterable object.
2. Explain the iteration protocol in Python.
3. What is the function of a generator?
4. What is the difference between yield and return?
5. What are list comprehensions? Explain how comprehensions help in

generating a sequence.
6. Explore some iteration tools in Python.
7. Do you believe that iter improves the time complexity vis-a-vis for?
1. Write a generator that produces the terms of arithmetic progression.
2. For the above question write the corresponding iterator class.

3. Write a generator that produces the terms of a geometrical progression.
5. Write a generator that produces the terms of a harmonic progression.
7. Write a generator that produces all the prime numbers up to a given number.
9. Write a generator that produces all the Fibonacci numbers up to n.
11. Write a generator that produces all the Armstrong numbers up to n.
13. Write a generator that produces Pythagoras triples in the range (1, 20).
15. Write a generator that produces all the multiples of 6 up to the given
number.
17. Write a list comprehension that produces all the numbers that are multiple of
2 or 5.
18. Write a list comprehension that converts a list containing the temperature in
degrees Celsius to that in Fahrenheit.
19. Write a list comprehension that produces all the prime numbers.
20. Write a list comprehension that produces all the numbers which leave
remainder 1 when divided by 5.
21. Write a list comprehension that produces all the vowels of a given string.
22. Write a list comprehension that produces the fourth power of numbers of a
given list.
23. Write a list comprehension that produces the absolute values of numbers in a
given list.
CHAPTER
7
FILE HANDLING
• Understand the importance of file handling
• Appreciate the mechanisms of file handling in Python
• Learn various file access modes and the open function
• Understand various functions for file handling in Python
• Implement the concepts studied in the chapter
7.1 INTRODUCTION
The data types and control structures discussed so far will help us to accomplish
many simple tasks. The problem so far is that we have not been able to store the
data or the results obtained for future use. Moreover, at times the results
produced by a program are voluminous. In such cases it becomes difficult to
store data in the memory or even to read the data. In such cases file handling
comes to our rescue.
The reader will also appreciate the fact that the main memory is volatile. The
data produced by a program cannot be used for future endeavors. Many times it
is required to store the data for use in future. For example, if one develops a
student management system, the user should be able to retrieve the data as and
when required.
As we understand, the data is stored in the binary format in a disk. Therefore,

while storing data the format of the data should be taken care of. At the
programmer’s level, however, the data can be stored in files or in databases.
Databases store and manage related data. The ease of retrieval, the security and
the flexibility make databases one of the most important topics in computer
science. The concept of databases, their usage, and related issues constitute a
dedicated subject. This chapter only concentrates on file handling. A file can be
perceived as set of records where each record has some fields. The fields, in
turn, have certain bytes. Files, as discussed later, can have many formats. This
chapter concentrates on binary and text files. The two formats differ in the
representation of the end of the file and in the storage of standard data types. A
file may have certain permissions associated with it. For example, one may not
have write permissions for a file which is to be used by the operating system. For
that matter, a user may not even have read permissions for such files. Such
constraints need to be kept in mind while writing programs for file handling.
Python provides many functions to carry out operations related to file handling.
The creation of a file, writing data to a file, reading the data, appending
something to the file and standard directory operations are discussed in this
chapter. Moreover, to make things interesting the use of the above operations in
encryption has also been discussed.
The chapter has been organized as follows. The second section discusses the
general file handling mechanism. The third section discusses the open() function
and the various modes in which a file can be opened. The fourth section
discusses the functions for reading and writing to the file. This section also
introduces the functions to get and set the position of a cursor in a file. The
fourth section also discusses some important factions to carry out various tasks
and the fifth section briefly discusses the command line arguments. The last
section of this chapter concludes.
7.2 THE FILE HANDLING MECHANISM

In Python, files are accessed using the file objects. As a matter of fact, the file
objects help us to access not just normal disk files but can help us to accomplish
many other tasks involving other kinds of files, which are explained later in this
chapter.
The file handling mechanism in Python is simple. The file needs to be opened
first. As in, the file is hooked to an Object [1]. This is done with the help of the
open() function. The function takes the name of the file and the mode as its
arguments. In fact, the function can have three arguments. The third is discussed
in the next section. The open function returns an object of the file. The object
then uses the library functions to read the file, write into it or append it. Finally,
the memory space occupied by the object is freed using the close() function.
The mechanism has been depicted in the following figure (Figure 7.1).
FIGURE 7.1 File handling in Python
FIGURE 7.2(a) The open function
FIGURE 7.2(b) The close function
Having discussed the mechanism of handling a file, now let us move on to the
file access modes and the open function in Python.
7.3 THE OPEN FUNCTION AND FILE ACCESS MODES

The files are accessed using the object created with the help of open() function.
In fact, there are many more functions used to create an object of the file type.
Note that the said functions return a file object or a file like object. This
abstraction is helpful for considering files as interfaces for communication. This
communication can be perceived as a transfer of bytes and the file can be
considered as a sequence of bytes.
So, in order to be able to do input/output to/from a file, the open() function is

needed. If the file is opened successfully, the file object is returned. If the file is
not opened successfully, the IOERROR exception is raised.
The open function takes three arguments. The first argument is the name of the
file, the second the mode in which the file is opened and the third indicates the
buffer string. As a matter of fact, the third will rarely be used. The first argument
is a string of characters, which is either a valid filename or a path. The path can
be relative or absolute. The access mode is the mode in which the file will be
opened (Figure 7.2). The various modes have been presented in Figure 7.3. The
modes open the file in read, write or append mode. In the read mode (“r”) the
file is opened, if it exists. The write mode (“w”) opens the file for writing. If the
file already exists, the existing contents of the file are truncated. The append
mode (“a”) opens the file for writing but does not truncate the existing contents.
In this mode if the file does not exist, it will be created.
FIGURE 7.3 File opening modes in Python
The modes can be suffixed with a letter “b” indicating binary access. The “+”
suffix can be used to grant read and write access to the file. Table 7.1 presents
the various modes and the corresponding operations that can be performed.
Table 7.1 Access modes for File

File Mode Operations
r reading from a file
w write to a file; creates the file if it does not exist; truncate the
file if it already exists.
a append to the file; if the file does not exist this creates the file
r+ open for read and write
w+ w for both read and write
a+ a for both read and write
rb read a binary file
wb write mode for a binary file
ab append mode for a binary file
rb+ r+ for a binary file
wb+ w+ for a binary file
ab+ a+ for a binary file
7.4 PYTHON FUNCTIONS FOR FILE HANDLING

Python provides various library functions to carry out the standard tasks. The
functions help us, say, to read from a file, write to a file and to append something
in the existing file. Moreover, Python also provides the programmer with
functions to take the cursor to a particular location, or to read from a given
location.
7.4.1 The Essential Ones

This section briefly introduces the various functions. The use of these functions
has been explained in the following sections. The reader is expected to
experiment with the functions in order to get a clear insight into them.
The read() function
The function reads bytes in a string. It may take an integer argument indicating
the number of bytes to read. If the argument is -1, the files must be read to the
end. Also if no argument is given, the default is taken as -1.
Tip
read() is same as read(-1)
If the content of the file is larger than the memory then only the content which
can fit into the memory will be read. Moreover, when the read operation ends a “
“(an empty string) is returned.
readline() and readlines()
The readline() method is used to read a line until the newline character is read.
It may be stated here that the newline character is retained in the string that is
returned. The readlines() method reads all the lines from a given file and
returns a list of strings.
write() and writelines()
The write() method writes the string in a given file. The method is
complementary to the read() method. The writelines() method writes a list of
strings to the file.
Tip
There is no writeline() method in Python 3.x
seek()
The seek() method takes the cursor to the starting position in the given file. The
position is decided with respect to the offset given. The offset can be 0, 1, or 2.
“0” indicates the beginning of the file. The value “1” indicates the current
position and the value “2” indicates the “end of the file.”
tell()
tell() is complementary to the seek() function. The function returns the

position of the cursor.
close()
The close() function closes the file. The object should be assigned to another
file after it is closed. Though Python closes a file after a program finishes (see
garbage collection in the following chapters), it is advisable to close the file
when the required task is accomplished. The repercussions of not closing the file
can be observed at the most unexpected times.
fileno()
The fileno() function returns a descriptor for the file. For example, the
descriptor of the file named “Textfile.txt,” in the following snippet is 3.
>>> f=open('Textfile.txt')
>>> f.fileno()
3
>>>
7.4.2 The OS Methods

The methods that deal with the issues related to operating systems help the
programmer to create a generic program. The methods also spare the
programmer from the horror of dealing with uncanny formatting details. For
example, the end of a line is represented by different character sets in different
operating systems. In Unix, a newline is indicated by “\n,” In MAC the newline
character is, “\r” and in DOS it is “\r\n.” Similarly, file separator Unix is “/,”,
whereas that in Windows is “\” and that in MAC is “:.” Their inconsistencies
make the life of a programmer miserable. This is the reason why a consistent
approach is needed to handle such situations. Table 7.2 presents the names and
functions of OS methods.
Table 7.2 OS methods
os method Function
linesep string used to separate lines in a file
sep used to separate file pathname components
pathsep delimit a set of file pathnames
curdir current directory
pardir parent directory
7.4.3 Miscellaneous Functions and File Attributes

As well as the functions stated above, flush and isatty are also used to make a
program more robust.
flush(): The flush function flushes the internal buffer.

isatty(): The function returns a “1,” if the file is a tty-like device.
For more such functions, the reader may refer to the Appendix of this book.
File attributes
It may also be stated here that the file attributes help the programmer to see the
state of a file and its features like the name, mode, and the softspace. Table 7.3
presents some of the most important file attributes.
Table 7.3 File attributes
File Attribute Importance

file.closed 1 if file is closed, 0 otherwise
file.mode access mode
file.name name of the file
The following illustration demonstrates the use of the above attributes.
Illustration 7.1: Open a file called “Textfile.txt” in the read mode. Check the
name of the file, its mode, and find whether it is closed using the file attributes.
Solution:
f=open('Textfile.txt','r')
print('Name of the file\t:',f.name)
print('Mode\t:',f.mode)
print('File closed?\t:',f.closed)
f.close()
print('Mode\t:',f.mode)
print('File cloased?\t:',f.closed)
Output
>>>
========= RUN C:/Python/file handling/fileattar.py =========
Name of the file : Textfile.txt
Mode : r
File closed? : False
Mode : r
File closed? : True
>>>
7.5 COMMAND LINE ARGUMENTS

If the compiler knows the name of the script, then the name of the script along
with the additional arguments that may be given along are stored in a list called
argv. The argv variable is in the sys module. The arguments, along with the
name of the script, are called the command line arguments. It may be noted here
that even the name of the script is part of the list. As a matter of fact, the name of
the script is the first element of the list. The rest of the arguments are stored in
the succeeding locations of the list. The argv can be accessed by importing the
sys module. The following illustration demonstrates the use of the argv variable.
Illustration 7.2: Display the number of command line arguments and the
individual arguments.
Solution:
import sys
print('The number of arguments',len(sys.argv))
print('Arguments\n')
for x in sys.argv:
print('Argument\t:',x)
Output
>>>
======= RUN C:/Python/file handling/commandLine.py =======
The number of arguments 1
Arguments
Argument : C:/Python/file handling/commandLine.py
>>>
The following example presents the bubble sort which takes the numbers entered
at the command line as the input.
Illustration 7.3: Sort the numbers (using bubble sort) entered as the command
line arguments.
Solution:
7.6 IMPLEMENTATION AND ILLUSTRATIONS

Having seen the mechanism of file handling, the functions and the attributes, let
us now have a look at the usage of the above functions. We will begin with the
most basic tasks and then use the functions to write something to a file (say
“TextFile.txt”), and open the file in the write mode. The open function, in this
case, will have two parameters: name of the file (“TextFile.txt”) and the mode
(“w”). Also, the file needs to be closed. Note that the write function returns the
number of bytes written in the file.
>>> f = open('TextFile.txt','w')
>>> f.write('Hi there\nHow are you?')
21
>>> f.close()
The read function reads the bytes of the given file. The open function, as stated
earlier, may not take any argument. This implies reading a file until the end. The
read text can be stored in a string (“text”).
>>> text=f.read()
>>> text
'Hi there\nHow are you?'
>>> f.close()
>>>
A file can be renamed using the rename function of OS. The rename function
takes two arguments: the first being the name of the original file and the second
being the new name of the file. In the following snippet, a file called
“TextFile.txt” is renamed to “TextFile1.txt” and read into “str” using the open
function.
>>> import os
>>> os.rename('TextFile.txt','TextFile1.txt')
>>> f=open('TextFile1.txt','r')
>>> str=f.read()
>>> str
'Hi thereHow are you'
>>>
Writing a list of strings in a file
As stated earlier, a list of strings can be written into a file using the
writelines() function. The use of the function has been illustrated as follows.
In the following snippet, the lines entered by the user are put into a list, L, and
this list is then written into the file f.
Illustration 7.4: Write a program to ask the user to enter lines of text. The user
should be able to enter any number of lines. In order to stop, he must enter “\e.”
The lines should be appended to an empty list (say L). This list should then be
written to a file called lines.txt. The program should then read the lines of
lines.txt.
Solution:
print('Enter text, press \'\\e\' to exit')

L=[]
i=1
in1=input('Line number'+str(i)+'\t:')
while(in1 !='\e'):
L.append(in1)
i=i+1
in1=input('Line number'+str(i)+'\t:')
print(L)
f=open('Lines.txt','w')
f.writelines(L)
f.close()
f=open('lines.txt','r')
for l in f.readline():
print(l, end=' ')
f.close()>>>
Output
========== RUN C:/Python/file handling/Write.py ==========
Enter text, press '\e' to exit
Line number1 :Hi there
Line number2 :How are you
Line number3 :I am good
Line number4 :\e
['Hi there', 'How are you', 'I am good']
Hi there How are youI am good
>>>
Reading n characters and the seek() function
The use of the read(n) function, which reads the first “n” characters of the file
has been demonstrated in the following illustration (Illustration 7.5). Note that
the tell function tells the position of the cursor, which is why the value of pos
changes as and when we move forward. The seek() function takes two
parameters, the first being the offset and the second the position. Note that
seek(0, 0) positions the cursor at the first position from the beginning.
Illustration 7.5: Open a file TextFile.txt and write a few lines in it. Now open
the file in the read mode and read the first 15 characters from the file. Then read
the next five characters. In each step show the position of the cursor in the file.
Now, go back to the first position in the file and read 20 characters from the file.
Solution:
f=open('TextFile.txt','w')
f.writelines(['Hi there', 'How are you'])
f.close()
f = open('TextFile.txt', 'r+')
str = f.read(15)
print('String str\t: ', str)
pos = f.tell()
print('Current position\t:', pos)
str1=f.read(5)
print('Str1\t:',str1)
pos = f.seek(0, 0)
print('Current position\t:',pos)
str = f.read(20);
print('Again read String is : ', str)
f.close()
Output
>>>
========= RUN C:\Python\file handling\Position.py =========
String str : Hi thereHow are
Current position : 15
Str1 : you
Current position : 0
Again read String is : Hi thereHow are you
>>>
Creating directories and navigating between them
One can also create directories in Python using the mkdir() function. The
function takes the name of the directory as one of the essential arguments. The
reader is advised go through the appendix of this book for a detailed description.
The chdir() function changes the current directory and the getpwd() function
prints the name (along with the path) of the current working directory. The use of
these functions has been demonstrated as follows:
'>>> import os
>>> os.mkdir('PythonDirectory')
>>> os.chdir('PythonDirectory')
>>> os.getcwd()
'C:\\Python\\file handling\\PythonDirectory'
>>>
An example of encryption
The following illustration uses the ord(c) function which prints the ASCII value
of the character “c,” and that of the chr(n) function which returns a character
corresponding to the ASCII value n.
Illustration 7.6: Write “Hi there how are you” in a file called “TextFile.txt.”
Now, read characters from the file, one by one and write the character obtained
by adding k (entered by the user) to the ASCII value of the character. Also,
decrypt the string in the second file by subtracting “k” from the ASCII values of
the characters in the second file.
Solution:
Output
Enter a number4
Character H Ascii value : 72
Character i Ascii value : 105
Character Ascii value : 32
Character t Ascii value : 116
Character h Ascii value : 104
Character e Ascii value : 101
Character r Ascii value : 114
Character h Ascii value : 104
Character o Ascii value : 111
Character w Ascii value : 119
Character a Ascii value : 97
Character r Ascii value : 114
Character y Ascii value : 121
Character o Ascii value : 111
Character u Ascii value : 117
Lm$xlivi$ls{$evi$}sy
Character L Ascii value : 76
Character m Ascii value : 109
Character $ Ascii value : 36
Character x Ascii value : 120
Character l Ascii value : 108
Character v Ascii value : 118
Character l Ascii value : 108
Character s Ascii value : 115
Character { Ascii value : 123
Character v Ascii value : 118
Character } Ascii value : 125
Character s Ascii value : 115
Character y Ascii value : 121
Hi there how are you
>>>
7.7 CONCLUSION
File handing provides the user with the power of persistence. The user must be
equipped with the knowhow of the file access modes, the open(), close()
functions and the functions which help in reading a file and writing to it. The
chapter briefly explains the most essential functions used for file handling in
Python. The chapter also introduces the user to OS methods and the essential file
attributes to help the user achieve the task at hand. The chapter includes ample
illustrations and explanations to make the concept clear in the simplest manner.
The reader is advised to go through the appendix for a detailed explanation of
the different types of files in Python and a detailed write up on command line
arguments.
POINTS TO REMEMBER
The open function takes three arguments.
The mode of opening file decides the tasks that can be accomplished.
The file should be closed after the required task has been completed.
The seek method helps to move the cursor within a file.
The file name attribute prints the name of the file.
The file mode attribute gives the file access mode.
The os.getpwd function returns the present working directory.
The os.chdir function changes the directory.
EXERCISES
1. Which of the following is a solid argument for using file handling?

(a) It is not possible to store all data produced by the program in the main
memory
(b) It is used for persistent storage
(c) Both
2. In which of the formats is the end of the line denoted by “\n” and “\r”?
(a) Text
(b) Binary
(c) Both
3. To be able to use a file it must be opened. The reason for doing so is

(a) To allocate memory to the object formed
(b) To specify the access mode
(c) To specify the offset (optional)
4. In f = open(“abc.txt”, “r”), the offset is

(a) 0 from the beginning
(b) 0 from the end
(c) Random
5. How many arguments does the open function take?

(a) 1
(b) 2
(c) 3
6. The file must be closed if it is opened in which of the following modes?

(a) r
(b) w
(c) Both
7. If the file is not opened successfully, which of the following exceptions is

raised?
(a) File not found
(b) IOERROR
(c) IO
8. In f = open(“abc.txt”, “w”), if the file “abc.txt” does not exist, then

(a) IOERROR is raised
(b) The program does not compile
(c) A new file is created
9. Which suffix is used for opening a binary file

(a) b
(b) bin
(c) ab
10. The + suffix allows

(a) Read
(b) Read and write
(c) Read or write
11. How many file access modes are there in Python?

(a) 3
(b) 6
(c) 9
(d) 12
12. The integer argument in the read() function denotes the number of bytes to
be read; if no argument is given, which of the following is the default
argument?
(a) –1
(b) 0
(c) len(file)
13. To read all the lines in a file, which of the following functions can be used?
(a) readline ()
(b) readlines ()
(c) Both
14. Which of the following methods can be used to write a list of strings in a
file?
(a) writeline ()
(b) writelines ()
(c) write ()
15. Which of the following arguments in the seek() function denotes the end of
the file?
(a) 1
(b) 2
(c) 0
16. Which function returns the descriptor of the file?

(a) fileno ()
(b) filedisp()
(c) descriptor ()
17. The linesep function is used to find which of the following?

(a) The new line
(b) The end of the file
(c) The current directory
18. Which of the following is not a file attribute?

(a) Closed
(b) Opened
(c) Name
(d) Softspace
19. In which of the following variables is the command line argument saved?
(a) argv
(b) argc
(c) Both
20. Which of the following functions helps to create a directory?

(a) os.mkdir()
(b) os.chdir()
(c) os.getpwd()
21. Which of the following functions helps to change the current directory?
(a) os.mkdir()
(b) os.chdir()
(c) os.getpwd()
22. Which of the following functions helps to print the name of the current
directory?
(a) os.mkdir()
(b) os.chdir()
(c) os.getpwd()
23. Which function is used to find the ASCII value of a character?

(a) ascii
(b) ord
(c) chord
24. Which of the following is not a file access mode in Python?

(a) a
(b) ab
(c) ab+
(d) abc
25. Which of the following is incorrect?

(a) f = open (‘file.txt’)
(b) f = open(‘file.txt’,’r’)
(c) f = open (‘file.txt’,’r’,0)
(d) None of the above is incorrect
THEORY
1. What is the importance of file handling? Explain the mechanism of file
handling in Python.
2. Explain various file access modes.
3. Explain the signature and usage of the following functions

(a) open
(b) close
(c) read
(d) write
(e) readline
(f) readlines
(g) writeline
(h) seek
4. What are file attributes? Explain the file attributes provided by Python.
5. Briefly explain the usage of the following os functions in Python

(a) mkdir
(b) chdir
(c) getpwds
PROGRAMMING
1. Write a program to copy the contents of one file to another.
2. Write a program to capitalize the first character of each word in a file.
3. Write a program to find the ASCII value of each character in a file.
4. Write a program to find the frequency of each character in a file.
5. Write a program to find all occurrences of a word, entered by the user, in a

given file.
6. Write a program to replace a given character with another in a file.
7. Write a program to replace a given word with another, in a given file.
8. Write a program to find the frequency of a given word in a file.
9. Write a program to find the word used the minimum number of times in a
given file.
10. Write a program to change the name of a file to the name entered by the
user.
11. Write a program to create a directory and then create a new file in it.
12. Write a program to print the name, number of characters, and number of
spaces in a file.
13. Write a program to convert the characters of a given file to binary format.
14. Write a program to find the words starting with a vowel from a given file.
15. Write a program to implement any substitution cipher on the text of a given
file.
CHAPTER
8
STRINGS
• Understand the concept and importance of strings
• Understand various string operators
• Learn about the built in functions to manipulate strings
• Learn how to solve problems using strings
8.1 INTRODUCTION
Strings are a sequence of characters. These data structures are used to store text.
For example if one wants to store the name of a person, or for that matter his
address, then string is the most appropriate data structure. As a matter of fact, the
knowledge of strings is essential in the development of many applications like
word processor and parser.
Strings in Python can be enclosed in single quotes or double quotes, or even in

triple quotes. However, there is no difference between a string enclosed in single
quotes or double quotes. That is “harsh” is same as “harsh.” Triple quotes are
generally used in special cases discussed later in the chapter. Strings in Python
come with a wide variety of operators and built-in functions.
The chapter examines various aspects of strings like non-mutability, traversal,

operators, and built-in functions. One of the most prominent differences between
a string and a list is non-mutability. Once a value is given to a string, one cannot
change the value of a character present at a particular position. For the users
familiar with C, C++, C#, or Java, the operators discussed in the chapter -
notably the * operator - will be a pleasant surprise. Moreover, Python provides
many built in functions to help the programmers to handle strings.
This chapter examines the above issues and provides examples of them. The
chapter has been organized as follows. The second section of the chapter
explores the use of standard 'for' and 'while' loops in strings. The third
section deals with the operators that can be used with strings. The built-in
functions used for accomplishing various tasks have been dealt with in the fourth
section and the last section concludes the chapter.
8.2 THE USE OF “FOR” AND “WHILE”
The traversal of a string has already been discussed in Chapter 4 of this book.
This section revisits the 'for' and 'while' and their applications in strings.
As stated in the second and the fourth chapter, strings are iterable objects. The
standard loops (read 'for' and 'while') can be used to iterate over a string. The
'for' loop helps to iterate through each character by storing the character in
some variable. The following illustration depicts the use of a for loop to iterate
the string.
The examples that follow use the for loop to carry out some basic and some
intricate tasks. Basic tasks like calculating the length of a given string have been
exemplified in Illustration 8.2. Illustrations 8.3, 8.4, and 8.5 implement
transposition and substitution.
Illustration 8.1: Write a program to traverse a string.
Solution: Writing for i in <string> helps us to access one character at a time

from a given string. The variable 'str1' stores the string entered by the user and
it is iterated using the for loop.
Listing
str1= input('Enter a string\t:')
for i in str1:
print('Character \t:',i)>>>
Output
=============== RUN C:/Python/String/str2.py ===============
Enter a string :harsh
Character :h
Character :a
Character :r
Character :s
Character :h
The above methodology can also help us to find the length of string. Note that
there is a built-in function to accomplish the said task. However, the purpose
here is to be able to use the 'for' loop in order to imitate the len function. In
the following illustration, a variable called length is initialized to 0 and is
incremented as we proceed.
Illustration 8.2: Write a program to find the length of the string entered by the
user.
Solution: The concept has already been explained in the above discussion. The
code follows.
Listing
name=input('Enter your name\t');
length=0
for i in name:
length=length +1
print('The length of ',name,' is ',length)
Output
=============== RUN C:/Python/String/str1.py ===============
Enter your name harsh
The length of harsh is 5
>>>
The ability to handle each character individually in a string gives the power to
manipulate a given string. One of the exciting tasks can be to implement basic
cryptography techniques. The example that follows displaces the characters two
positions to the right. This is referred to as transposition. The next example shifts
the characters by “k” positions, “k” being entered by the user.
Illustration 8.3: Ask the user to enter a string and displace two characters to the
right.
Solution: Note that in each iteration, the position of the characters is shifted by
two positions. The code follows:
str1=input('Enter the string\t:')
i=0
str2=""
while i<len(str1):
str2[i]=str1[(i+2)%len(str1)]
print(str2)
Illustration 8.4: Ask the user to enter a string and displace k characters to the
right.
Solution: Note that in each iteration, the position of the characters is shifted by k
positions. The code follows:
k=int(input('Enter the value of k\t:'))
i=0
str2=""
while i<len(str1):
str2+=str1[(i+k)%len(str1)]
print(str2)
i+=1
print(str2)
>>>
Output
========== RUN C:/Python/String/transposition.py ==========
Enter the string :harsh
Enter the value of k 4
h
hh
hha
hhar
hhars
hhars
>>>
Another method of encryption is substitution. The replacement of a symbol by

some other symbol is referred to as substitution. The example that follows
implements one of the most basic types of substitutions. Here, each character is
replaced by a character obtained by adding two to the ASCII value of the
character, and finding the requisite character.
Illustration 8.5: Ask the user to enter a string. Replace each character by that
obtained by adding two to the ASCII value of that character.
Solution:
k=int(input('Enter the value of k\t:'))
i=0
str2=""
while i<len(str1):
str2+=str((ascii(str1[i])+k))
print(str2)
i+=1
print(str2)
8.3 STRING OPERATORS

Python provides the programmer with a wide variety of extremely useful
operators to manipulate strings. These operators help a user perform involved
tasks with ease and efficiency. Here, it may be stated that the replication and
membership operators make Python stand apart from its counterparts. This
section briefly introduces and exemplifies these operators.
8.3.1 The Concatenation Operator (+)

The concatenation operator takes two strings and produces a concatenated string.
The operator acts on values as well as variables. In the examples that follow, the
concatenation operator’s result has been stored in variables called result1 and
str2.
name=input('Enter your name\t:')

result1 = 'Hi'+' there'
print(result1)
str1='Hello'
str2=str1 +' '+name
print(str2)
Output
>>>
============= RUN C:/Python/String/operator1.py
=============
Enter your name :Harsh
Hi there
Hello Harsh
>>>
Note that the same operator is used for adding two integers.
8.3.2 The Replication Operator

The replication operator in Python replicates the strings as many times as the
first operand. The operator operates on two operands: the first being a number
and the second being a string. The result is a string in which the input string is
repeated as many times as the first argument. In the example that follows, the
result has been stored in a variable called result1.
name=input('Enter your name\t:')
print('Hi', ' ', name)
str1=input('Enter a string\t:')
num=int(input('Enter a number\t:'))
result1=num*str1
print(result1)
Output
>>>
============= RUN C:/Python/String/operato2.py =============
Enter your name :harsh
Hi harsh
Enter a string :abc
Enter a number 4
abcabcabcabc
>>>
8.3.3 The Membership Operator

The membership operator checks whether a given string is present in a given list
or not. The operator returns a True if the first string is a part of the given list,
otherwise it returns a False.
>>> 'Hari' in ['Har', 'Hari', 'Hai']
True
>>>
>>> 'Hari' in ['Har', 'hari', 'Hai']
False
>>>
It may be noted here that this operator is also used for manipulating iterations.
The reader is advised to go through Chapter 4 of this book for a detailed
discussion regarding the use of “in” in for. It may also be noted that the operator
can also be used in tuples. In the listing that follows, the string “Hari” is present
in the given tuple and hence True is returned.
>>> 'Hari' in ('Hari', 'Har')
True
>>>
The reader may also note that corresponding to the “in” operator, there is a “not
in” operator which works in the exactly opposite manner vis-a-vis “in.”
A string in Python can span over many lines. This can be accomplished by
putting a “\” at the end of the line. For example, str2 is "Harsh Bhasin Author
Delhi". However, it has been written in three lines using the “\” character.
>>> str2="'Harsh Bhasin\

Author\
Delhi'"
>>> str2
"'Harsh BhasinAuthorDelhi'"
8.4 FUNCTIONS FOR STRING HANDLING

This section presents some of the most common functions used to manipulate
strings in Python. It may be stated here that, although all the following tasks can
be done without the predefined functions with varying degree of ease, the
presence of these functions help the programmer to do the task easily and
efficiently. Moreover when one crafts and implements one’s version of a
function, the implementation may not be efficient in terms of time or space or
both. However, while crafting these predefined functions in Python the issues
related to memory and time should be handled. Let us now have a look at the
names, meanings, and usage of the pre-defined functions in Python.
8.4.1 len()
Usage:
>>> len()
Explanation:
The function returns the number of characters in a string. For example if a

variable called str1 stores 'Harsh Bhasin', then the length of the string can be
calculated by writing len(str1). Note that the space between 'Harsh' and
'Bhasin' has also been taken into account while calculating the length of the
string. The function takes a string argument and returns an integer, which is the
length of the string.
Example (s):
>>> str1 ='Harsh Bhasin'
>>> len(str1)
12
>>>
>>> len('Harsh Bhasin')
12
>>>
>>> len('')
0
8.4.2 Capitalize()
Usage:
>>> capitalize()
Explanation:
The function capitalizes the first character of the string. Note that only the first
character will be capitalized. If one wants to capitalize the first characters of all
the words in the string the title() function can be used.
Example (s):
>>> str2='harsh bhasin'
>>> str2
'harsh bhasin'
>>> str2.capitalize()
'Harsh bhasin'
8.4.3 find()
Usage:
>>><name of the string>.find(<parameter(s)>)
Explanation:
The location of a given sub-string in a given string can be found by using the
function find. Also, if the location of a sub-string after a particular position (and
before a particular index) is to be determined, then three arguments can be
passed to the function: the sub-string, initial index, and the final index. The
following examples show the usage of the function.
Example(s):
>>> str2.find('ha')
0
>>>
>>> str2.find('ha',3,len(str2))
7
8.4.4 count
Usage:
>>><name of the string>.count(<parameter(s)>)
Explanation:
The number of occurrences of a particular substring can be found with the count
function. The function takes three arguments: the sub-string, the initial index,
and the final index. The following examples show the usage of the function.
Example(s):
>>> str3.count('ha',0,len(str3))
1
>>> str3.count('ka',0,len(str3))
0
8.4.5 Endswith()
Usage: <name of the string>.endswith(<parameter(s)>)
Explanation:
One can determine if a string ends with a particular sub-string. This can be done
using the endswith() function. The function returns a 'True' if the given string
ends with the given sub-string, otherwise it returns a 'False'.
Example(s):
>>> str3.endswith('n')
True
8.4.6 Encode
Usage: <name of the string>.encode(<parameter(s)>)
>>>
Explanation:
Python provides a function to encode a given string in various formats. The

function is encode. It takes two arguments: encoding=<value> and errors=
<value>. The encoding can be one of the many encodings given in Appendix B
of this book. The following examples demonstrate the use of this function.
Example(s):
>>> str3.encode(encoding='utf32',errors='strict')
b'\xff\xfe\x00\x00H\x00\x00\x00A\x00\x00\x00R\x00\x00\
x00S\x00\x00\x00H\x00\x00\x00\x00\x00\x00b\x00\x00\
x00h\x00\x00\x00a\x00\x00\x00s\x00\x00\x00i\x00\x00\
x00n\x00\x00\x00'
8.4.7 Decode
Usage:
>>><name of the string>.decode(<parameter(s)>)
Explanation:
The function returns the decoded string.
8.4.8 Miscellaneous Functions

Though the purpose, usage and examples of the most important functions have
been explained, the following functions are also important. A list of some more
functions follows. The list is followed by a brief explanation.
List:
1. isanum()
2. isalpha()
3. isdecimal()
4. isdigit()
5. isidentifier()
6. islower()
7. isupper()
8. swapcase()
9. isspace()
10. lstrip()
11. rstrip()
12. replace()
13. join()
Explanation:
The contents of a given string can be checked using the following functions. The
isalnum() function if the given string is alphanumeric. The other functions like
isalpha() and isdecimal() also check the type of contents in a given string.
Whether a given string contains only digits can be checked using the isdigit()
function. Similarly, whether a given string is an identifier can be checked using
the isidentifier() function. The islower() function checks if the given string
contains only lower case characters and the isupper() function checks if the
given string contains only upper case characters. The swapcase() function
swaps the case of the given string, as in converts the upper case to lower and
lower to upper. The presence of spaces (only) can be checked using the
isspace() function. Extra spaces can be removed from the left and the right
hand sides by using the lstrip() and rstrip() functions. The replace()
function replaces the instances of the first argument by the string in the second
argument. The split function splits the given strings into tokens. Illustration 8.6
depicts the use of this function for splitting the string into constituent words. The
function of join() is exactly the opposite as that of split.
Example(s):
>>> str3.isalnum()
False
>>> str3.isalpha()
False
>>>
>>> str3.isdecimal()
False
>>>
>>> str3.isdigit()
False
>>>
>>> str3.isidentifier()
False
>>>
>>> str3.islower()
False
>>>
>>> str3.isnumeric()
False
>>>
>>> str3.replace('h','p')
'HARSH bpasin'
>>>
Illustration 8.6: A string str4 contains a sentence “I am a good boy.” Split the
string and also display each token using a for loop.
Solution:
>>> str4='I am a good boy'
>>> str4.split()
['I', 'am', 'a', 'good', 'boy']
>>>
>>> for i in str4.split():
print('Token\t:',i)
Output
Token : I
Token : am
Token : a
Token : good
Token : boy
8.5 CONCLUSION
In C and C++, strings used to be character arrays. They were a special type of
arrays with a “\0” character at the end. Strings in “C” came with a set of built-in
functions. However, there were two problems. Firstly string was not an
independent data type, and secondly an individual character could be changed.
In Python, the importance of strings has been duly recognized by creating an
object type. Moreover strings, in Python, are non-mutable. Strings come with a
wide range of built-in functions. Also there are useful operators to help the
programmer accomplish a given task easily and efficiently. This chapter
introduces the concept, operators and functions of strings. However, the reader is
expected to complete the end-chapter exercise to be able to understand and use
strings.
GLOSSARY
String: Strings are a sequence of characters. These data structures are used to
store text.
IMPORTANT POINTS
Strings in Python are non-mutable.
The negative index denotes the characters from the right hand.
Strings are iterable objects.
EXERCISES

(a) A string in Python is iterable
(b) A string in Python is not iterable
(c) Iterablity of a string depends upon the situation
2. Is a string in Python mutable?

(a) No
(b) Yes
(c) Depends on the situation
3. If str1='Hari', what is the output of print(str1[4])

(a) i
(b) \0
(c) Exception is raised
4. If str1='Hari', what is the output of print(str1[-3])

(a) “a”
(b) “H”
5. What is the output of "Hari"=="hari"

(a) True
(b) False
(c) An exception is raised
6. What is the output of 'a'<>'A'

(a) True
(b) False
7. What is the output of '567'>'989'

(a) True
(b) False
(c) An exception is raised
8. Which of the following helps to find the ASCII value of “C”?

(a) ord('C')
(b) chr('C')
(c) both
9. Which of the following helps to find the character represented by ASCII

value 67?
(a) ord(67)
(b) chr(67)
(c) Both
10. What are 'in' and 'not in', in Python?

(a) Relational operators
(b) Membership operators
(c) Concatenation operators
11. What is the output of 'A' + 'B'

(a) 'A+B'
(b) 'AB'
(c) 131
12. What is the output of 3*'A'

(a) '3A'
(b) Character corresponding to the ascii value 65 × 3?

(c) 'AAA'
13. Which function capitalizes the first character of a given string?

(a) capitilize()
(b) titlecase()
(c) toupper()
14. The find() function in Python takes

(a) 1 argument
(b) 3 arguments
(c) Both
15. If str1='hari', then what would be the output of str1.asalnum()?

(a) True
(b) False
16. If str1='hari3', then what would be the output of str1.asalnum()?

(a) True
(b) False
17. If str1='hari feb', then what would be the output of str1.asalnum()?

(a) True
(b) False
18. If str1='123h', then what would be the output of str1.digit()?

(a) True
(b) False
19. Which function checks whether all the characters in a given string are in
lower case?
(a) lower()
(b) islower()
(c) istitle()
20. Which function checks whether all the characters in a given string are in
upper case?
(a) upper()
(b) isupper()
(c) istitle()
21. Which function removes the whitespaces from the right hand of a given
string?
(a) rstrip()
(b) strip()
(c) lstrip()
22. Which of the following functions convert a given string into a list of words?
(a) split()
(b) break()
(c) breakup()
23. Which of the following helps in breaking a string into two substrings of
desirable length?
(a) Slicing
(b) Splitting
(c) Both
24. Which of the following functions combines the strings given as the
argument?
(a) Split
(b) Join
(c) Slice
25. Which of the following is illegal in Python (assume that str1 is a string,
having initial value 'hari')?
(a) str1= 'Harsh'
(b) str1[0]= 't'
(c) str1[0]=str[2]
THEORY
1. What is a string? Explain non-mutability. Is there any difference between a
string in double quotes and that in triple quotes?
2. Explain the following functions vis-vis strings.

+
*
in
not-in
3. Explain the following string operators by giving an example.

capitalize()
title()
len()
find()
count()
endswith()
encode()
decode()
4. What is the difference between a string in Python and a string in “C”?
5. What is the difference between a list and a string?
PROGRAMMING PROBLEMS
1. Write a program to reverse a string.
2. Write a program to encode a string in UTF format.

3. Write a program to find the sum of ASCII values of the characters of a given
string.
4. Write a program to find a particular substring in a given string.
5. Write a program to split a given text into tokens.
6. Write a program to check which of the tokens obtained in the above question
are keywords.
7. Write a program to check how many alphanumeric strings there are in the
tokens obtained in question 5.
8. Write a program to check how many alpha strings there are in the tokens
obtained in question 5.
9. Write a program to check how many numeric strings there are in the tokens
obtained in question 5.
10. Write a program to convert a string entered by a user to that obtained by

adding “k” to each character’s ASCII value.
11. Implement the first phase of compiler design (for “C”). Please refer to the
following link for a brief overview of compiler design.
12. In the above question design deterministic finite acceptors for the keywords.
USEFUL LINKS
Strings and regular expressions are extensively used in developing the first phase
of a compiler. The following links will be helpful to understand the topic:
https://www.cs.cmu.edu/~fp/courses/15411-f13/lectures/07-lex.pdf
http://www.iith.ac.in/~ramakrishna/Compilers-Aug15/slides/02-lexical-
analysis-part-1.pdf
https://www.cs.utexas.edu/users/novak/cs375contents.html
CHAPTER
9
INTRODUCTION TO OBJECT
ORIENTED PARADIGM
• Understand procedural, modular and object oriented paradigm
• Understand the concept of class
• Design a class
• Understand the elements of object oriented programming
9.1 INTRODUCTION
In the preceding chapters, the control structures of Python were discussed. The
first section discussed loops, conditional statements, etc. However, these
constructs were an integral part of C as well, which is a procedural language. The
procedural programming is one that uses procedures. Each procedure is a set of
instructions where each instruction directs the computer what is to be done.
Python also supports object-oriented programming (OOP). This chapter
introduces the principles of OOP and explains the need and importance of
classes and objects. The chapter also discusses the difference between OOP and
procedural programming to give an insight of why OOP is needed.
It may be stated here that the topics discussed in this chapter will be discussed in
detail in the following chapters. Some of the readers not familiar with C++ (or
for that matter C# or Java) may find the discussion abstract, but things will
become clear as we proceed.
As stated earlier, in procedural programming each statement tells the program

what to do. For example the following code asks the user for the input, calculates
the square root and displays the result.
Code
>>> a = float(input("Enter a number\t:"))
Enter a number: 67
>>> b = math.sqrt(a)
>>> b
8.18535277187245
>>>
FIGURE 9.1 Programming paradigms
This strategy is good if the program is very small. Often telling the computer
what to do, step by step, works if the task to be accomplished is not very
complex. In such cases no other paradigm is needed.
In case of a moderately large program, division into functions makes the task
easier. The division of a larger program into functions makes the program
manageable and helps to achieve reusability of code. The functions generally
accomplish a clearly defined task and become handy whenever that particular
task is to be accomplished. The reader is advised to go through the chapter on
functions in order to understand the advantages of functions. The clubbing
together of functions on some basis give rise to what are commonly referred to
as modules. The programming paradigm is called modular programming.
The problem with this paradigm is that the accidental clubbing together of
unrelated functions, far from the real world situations, become a source of
problems at some point in time. Moreover, the approach does not restrict the
access of data in any module and may jeopardize the sanctity of the data.
It may be noted that the data should not be accessible to all the modules. The
accessibility of data must be managed with utmost care otherwise a module,
which should not have alerted the data as per the program logic, might change
the data.
In order to understand the gravity of the problem, let us take an example of C. In

C, a variable can be global or local. If it is global, then any module can change
it. If it is local, then other modules cannot access it. So there is nothing in
between. That is, we cannot make a variable which can be accessed only by
designated methods not data.
The solution to the above problem is to model the software in such a way that
the design is conceptually as close to the real world as possible. This modelling
of real world situations requires the creation of entities having both attributes
and behavior. The clubbing together of data and the functions that manipulate
the data are be helpful in crafting the above entities. These entities will
henceforth be referred to as classes. The instances of classes are objects and
the paradigm is called object oriented paradigm. Various programming
paradigms and their disadvantages have been summarized in Figure 9.1.
9.2 CREATING NEW TYPES

Though types are not explicitly declared in Python, the types were important in
other languages (well most of them). For example, when one says that a
“number” is of integer type, then he not only states the type of information but
also its maximum and minimum value. Assume that an integer takes two bytes;
the maximum value of “number” would be 32, 767 and the minimum value
would be –32, 768. Moreover, saying that “number” is of integer type also
specifies the operations that can be performed on the number.
Integer is a pre-defined type. Most of the languages also allow the user to create
custom types and hence extend the power of built-in types. This is essential as
the ability to create new data types will help us to create programs which are
near to the real world. For example if one has to design an inventory
management system then a type called “item” would make the matters
uncomplicated. This item can have variables which are of predefined types, like
integers and strings.
A new type can be created by declaring a class. A class has many components,
most important of which are attributes and functions. This clubbing together of
functions and data forms the basis of OOP. The functions, as we will see later,
generally manipulate the data members of a class. Before proceeding any further
let us have an overview of attributes and functions.
9.3 ATTRIBUTES AND FUNCTIONS

One can perceive a class as a prototype and an object as an instance of a class.
For example, "movie" is a class and "The Fault in Our Stars", "Love
Actually" and "Sarat" are objects (Figure 2). A class has attributes and
behavior. The attributes generally store data and the behavior is implemented
using functions. A class can be depicted using a class diagram. A class diagram
has, generally, three parts; the first part contains the name, the second part has
attributes, and the third part shows the functions of a class. The basics of
attributes and behavior are discussed in the following section. In Figure 9.2, the
class diagram (movie) has only the name.
FIGURE 9.2 Example of a class and objects
9.3.1 Attributes
The attributes here depict the characteristics of the entity that we are concerned
with. For example, when creating a website that gives the details of movies, a
class 'movie' will be needed. Say after detailed deliberations it was decided that
this class would have attributes like name, year, genre, director, producer, actors,
music director, and story writer.
Note that for the said purpose, only the above details are needed. Storing
unnecessary details will not only make data management difficult, but will also
violate one of the core principles - that of including only the details pertaining to
the problem at hand. These attributes are generally shown in the second section
of the class diagram. In Figure 9.3, the attributes of "movie" class have been
shown.
FIGURE 9.3 Name and attributes of a movie class
9.3.2 Functions
The next step will be to include functions in the above class. In our example
there are two functions - getdata() and putdata(). The getdata() function
asks for the values of the variables from the user and the putdata() function
will display the data. Functions implement the behavior of a class. The
functions, as stated earlier, accomplish a particular task. In a class there can be
any number of functions, each accomplishing a particular task. As a matter of
fact we have special functions for initializing the data members of a class as
well. The functions of a class will henceforth be referred to as member
functions. The functions (or behavior) are shown in the third section of a class
diagram. In Figure 9.4, the functions of the “movie” class (getdata()and
putdata()) have been shown in the third box.
FIGURE 9.4 Name, attributes, and functions of a movie class

The following example shows a class called movies. The class has the following
data members:
Name
Year
Genre
Director
Producer
Actors
Music_Director
Story_writer
The class has two functions - getdata(), which asks the user to enter the values
of the data members and putdata(), which displays the values of the variables.
In order to call the functions getdata() and putdata(), an instance of the
employee class is created ('m'). As we will see later, the functions are called
using the dot operator. The details regarding the syntax will be explained in the
following chapter.
The following code implements the above class. Though the syntax etic.. has not
been discussed as of yet, the code has been given to give an idea of how things
actually work.
Code
Output
============ RUN C:/Python/Class/class_basic2.py============
Enter name :Kapoor
Enter year :2016
Enter genre :Drama
Enter the name of the director :ABC
Enter the producer :Karan
Enter the name of the actor :Siddarth
Press 'y' for more 'n' to quity
Enter the name of the actor :Fawad
Enter 'y' for more 'n' to quitn
Enter the name of the music director :XYZ
Name :Kapoor
Year 2016
Genre :Drama
Director :ABC
Producer :Karan
Music_director :XYZ
Actors :['Siddarth', 'Fawad']
>>>
In object-oriented languages, a special function initializes the value of the data

members. This function generally has the same name as that of the class. The
function is called constructor.
One can create a default constructor in a class, which does not take any
parameters. The parameterized constructor, on the other hand, takes
arguments and initializes the data members using those arguments. The
implementation of constructors and their uses will be dealt with in the next
chapter.
When the lifetime of an object ends, a destructor is called. A destructor can be

called using 'del' in Python. The concept has been explained in the next chapter
of this book.
Tip
A constructor acts when an object is created and a destructor is called when the
lifetime of an object ends.
9.4 ELEMENTS OF OBJECT-ORIENTED PROGRAMMING

The following discussion briefly outlines the principles of object-oriented
programming. The concepts encapsulation, data hiding, and polymorphism are
all discussed in this section.
9.4.1 Class
A class is a real or a virtual entity, which has relevance to the problem at hand
and has sharp physical boundaries. A class can be a real entity. For example,
when one develops software for a car wash company, then 'Car', is central to
the software and therefore there will be a class called 'Car'. A class can also be
a virtual entity. Another example is that when developing a student management
system, a 'student' class is crafted which is a virtual entity. In both the
examples, the entity was crafted as it was important to the problem at hand.
The example of the 'student' class can be taken further. The class will have
attributes, which are needed in the program. The selection of attributes will
decide on the physical boundaries of the class. The fact of the matter is we will
not be needing unnecessary details like the number of cars a student has or
where he went last evening in an educational institute for which we are making
the student management system, so there is no point storing those details.
Examples of some of the classes that are central to the stated software are as
follows (Table 9.1).
Table 9.1 Examples of classes central in various systems
System class central to the software

Student Management System Student
Employee Management System Employee
Inventory Control Item
Library Management Book
Movie Review Movie
Airline Management Flight
Examination Test
9.4.2 Object
Consider a student management system that stores the data of each student of a
school. Note that while entering data the operator would deal with an individual
student, not the idea of a student. The idea of a student is a class, whereas each
student is an instance of class or an object.
An object is an instance of a class. The objects interact with each other and get
the work done. Generally, a class can have any number of objects. One can even
form an array of objects. The example of 'movie' had 'm' as an object. As a
matter of fact, we make an object and call the methods of a class (those which
can be called).
In object-oriented paradigm, the program revolves around an object and

therefore the type of programming is termed as an object-oriented program.
Calling a method of an object is equivalent to sending message to an object.
9.4.3 Encapsulation
The class is an entity, which has both data and functions. The clubbing together
of the data and the functions that operate on the data is called encapsulation.
Encapsulation is one of the core principles of object-oriented paradigm.
Encapsulation not only makes it easier to handle objects but also improves the
manageability of the software.
Moreover, the functions in a class can be used in variety of ways. Their

accessibility of data members and member functions can also be managed using
access specifiers, as explained in the following sub-section.
9.4.4 Data Hiding

Data hiding is another important principle of object-oriented programming. As
stated in the above discussion, the accessibility of data can be governed in a
class. The data, in the case of procedural programming, is accessible to all
throughout the program. This is referred to as global data. The data private to a
class is one which can be accessed only by the members of the class. There are
other access specifiers as well, explained in the following sections.
In C++, for example, the data in a class is generally kept private. That is, only
the member functions of the class can access the data. This ensures that the data
is not accidently changed. The functions, on the other hand, are public in C++.
The public functions can be accessed anywhere in the program (Figure 9.5). In
C++, Java, C#, etc., there is another access specifier, which is protected. If a
member is to be accessed in the class and the derived class, then a protected
specifier is used. C# and Java also have some other specifiers like internal.
FIGURE 9.5 Access specifiers, public and private
Having discussed the data access, it must be clarified that deciding what is
private and what is public is up to the discretion of the design and development
team of the project. There is no hard and fast rule as to what should be private
and what should be public. The designers must decide on the accessibility of a
member based on their needs.
This protection of data is not related to the security of data but to accidental
change. This is needed so that the data can be changed only via the functions
which have the authority to change data.
9.4.5 Inheritance
Classes are made so that they can be sub-classed. This art of dividing the class
into subclass(es) is inheritance. For example, the movie class can be sub-classed
into various classes like art_movie, commercial_movie, etc. Likewise, the
student class can be sub classed into 'regular student' and
'part_time_student'. In both the examples the subclass has many things which
are there in the base class (the class from which the sub-class has been derived)
in common. In addition, each subclass can have functions and data that belongs
to the sub-class only.
For example, the student class can have attributes namely name, date_of_
birth, address etc. The subclass regular student will use all the above data
members and can also have attributes like attendance associated with it. The
class from which classes will be sub-classed is referred to as the base class and
the subclasses would be called derived classes.
For example in Figure 9.6, movie is the base class and commercial_movie and
art_movie are the sub-classes.
FIGURE 9.6 Deriving classes from other classes is inheritance. There are many types of
inheritance. The above figure shows hierarchical inheritance
9.4.6 Polymorphism
Poly means many and morphism is forms, so polymorphism means many forms.
Polymorphism can be implemented in many ways. One of the simplest examples
of polymorphism is operator overloading. Operator overloading means using the
same operator in more than one way. For example “+” is used between integers
to add, with strings for concatenation and even can be used in user defined data
types as explained in Chapter 12 of this book.
Likewise function overloading means having more than one function with the
same name in a class with different arguments. Various forms of polymorphism
are explained in Chapters 10 and 11 of this book.
9.4.7 Reusability
The procedural programming came with almost no reusability. Modular
programming allowed reusability but only to certain extend. The functions could
be used on an “as is basis” in modular programming. In object-oriented
programming, the concept of reusability can be used in its full force. The
concept of inheritance, introduced above and explained in Chapter 10 of this
book, helps the programmer to reuse a code as per their requirements. a matter of
fact, reusability is one of the USPs of the object-oriented paradigm.
However, there is a catch. Lately, some researchers have cast doubts on the
ability of OOP vis-a-vis reusability.
9.5 CONCLUSION
While designing software, one must keep in mind the entities he is going to work
on. The nitty-gritty can be decided at a later stage. As a matter of fact, popular
literature does not consider the details of the operation as a matter of concern for
the object-oriented programming. Hiding unnecessary details are, therefore, an
important part of object-oriented programming.
For example, while developing a website for movies, the entity central to the
problem is “Movie.” So, one starts with an empty class called "movie". The
designer must then decide on the attributes needed to implement the functions.
The attributes constitute the data members of the said class. The behavior of the
entity is then deliberated upon. The member function determines the behavior of
a class. The functions are then designed. The things like inheritance and
polymorphism, discussed later in this section, come into play. And finally the
system is created.
This journey of the formation of a class has been depicted in the following figure
(Figure 9.7).
FIGURE 9.7 The design of a movie class
Programming is an art. A good programmer should be well-versed in the syntax

of the language, the data structures and the concepts of algorithm analysis. In
addition to the above, a programmer needs to decide the programming paradigm
that he is going to use. The chapter briefly introduces various programming
paradigms and the advantages and disadvantages of them. The chapter
introduces the concept of object-oriented programming. The definitions of class,
object, etc., have been discussed in the chapter. The chapter also introduces the
features of OOP. The concepts introduced in this chapter will form the
foundation of the rest of the chapter of this section. As already stated some of the
concepts may appear abstract at this stage, but the following chapters will revisit
the concepts and will demonstrate the implementation of the ideas dealt with in
this chapter. In order to be able to make a program that uses OOP, one must get
out of the mindset of doing things in procedural way and start thinking about the
program that is centered on real world entities having attributes and behaviors.
It may also be stated that the designing of an object-oriented program is

generally preceded by the design of class diagrams and sequence diagrams etc.
These are part of Unified Modelling Language. The concept of class diagrams
has been introduced in this chapter and UML has been introduced in the
references at the end of the chapter. The reader is advised to go through the
references at the end of the chapter before proceeding any further.
GLOSSARY
Class: A class is a real or a virtual entity that has relevance to the problem
at hand and has sharp physical boundaries.
Object: An object is an instance of a class.
Encapsulation: The clubbing together of the data and the functions that
operate on the data is called encapsulation.
Inheritance: The art of dividing the class into subclass(es) is inheritance.
Operator overloading: Operator overloading, in general, means using the
same operator in more than one way.
Function overloading: This means having more than one function with the
same name in a class with different arguments.
POINTS TO REMEMBER
Telling the computer what to do, step by step, works if the task to be
accomplished is not very complex. In such cases no other paradigm is
needed.
In the case of a moderately large program, division into functions makes the
task easier.
The division of a larger program into modules makes the program
manageable and helps to achieve reusability of code.
The clubbing together of functions, on some basis, gives rise to what is
commonly referred to as modules. The programming paradigm is called
modular programming.
A class has two important components: attributes and behavior.
A constructor initializes the members of a class.
The destructor frees the memory occupied by an object.
EXERCISES
1. Which of the following is not object-oriented language?

(a) C
(b) C++
(c) Python
(d) C#
2. Which of the following is object-oriented language?

(a) Python
(b) C#
(c) Java
3. A student is a conceptual entity, which acts as a blueprint for each student.

The mapping is similar to which of the following?
(a) Class and object
(b) Method and modular programming
(c) Both
4. Which of the following are the two most important components of a class?
(a) Methods and attributes
(b) List and tuple
(c) Arrays and functions
(d) None of the following
5. In object-oriented paradigm, a variable of a class is called

(a) Data member
(b) Member function
(c) Global data
6. In object-oriented paradigm, the functions of a class are

(a) Member functions
(b) Data members
(c) Global functions
7. An instance of a class is called

(a) Object
(b) Subject
(c) Inject
8. The clubbing together of data and the functions that operate on the data is
called
(a) Abstraction
(b) Encapsulation
(c) Overloading
9. Allowing the selective access of data members in a class is the same as
(a) Data Hiding
(b) Encapsulation
(c) Abstraction
10. If we have the same name functions in a class, then it is called

(a) Function overloading
(b) Overriding
(c) Encapsulation
11. “+” can be used for adding two number types. However, a programmer can
use “+” for the addition of two user defined data types (for example,
complex numbers).
This is
(a) Method overloading
(b) Operator overloading
(c) Encapsulation
12. Inheritance is helpful in handling

(a) Reusability
(b) Redundancy
(c) Overhead
13. If a function in the base class is extended in the derived class, then it is
(a) Overloading
(b) Abstraction
(c) Encapsulation
14. Which of the following is not a type of inheritance?

(a) Simple
(b) Multiple
(c) Hierarchical
(d) All of them are types of inheritance
15. Which of the following initializes the members of a class?

(a) Constructor
(b) Destructor
(c) Both
16. Which of the following is true for a well-defined class?

(a) It has importance to a problem at hand
(b) It has sharp physical boundaries
(c) It is a real or a physical entity
17. A language in which one can define a new data type is

(a) Comprehensive
(b) Extensible
(c) Both
18. In object-oriented paradigm, the focus is on

(a) Data
(b) Way a work is done
(c) Data Types
19. UML is
(a) Ultra-Modern Language
(b) Unified Modelling Language
(c) United Model League
20. Which of the following is not a principle of object-oriented paradigm?

(a) Inheritance
(b) Data hiding
(c) Encapsulation
(d) Divide and conquer
THEORY
1. Briefly explain the various paradigms of programming.
2. What is the difference between object-oriented paradigm and procedural

programming?
3. What is a class? What are the essential components of a class? Define

attributes and functions of a class.
4. What is the relation between an object and a class?
5. What is a class diagram? Give an example of a class diagram.
6. Explain the importance of encapsulation.
7. Explain the importance of data hiding. Is it related to the security of the

data?
8. What is polymorphism? Explain the concept of operator overloading and

function overloading.
9. What is the advantage of reusability? Explain the concept of reusability vis-

a-vis object-oriented paradigm.
10. Explore some of the problems in object-oriented programming?
EXPLORE AND DESIGN

The reader is expected to go through material on the subject: Database
Management System. The chapters on entity relationship diagrams have details
of entities involved therein. Create class diagrams of the classes mentioned in
Table 9.1, based on your research.
CHAPTER
10
CLASSES AND OBJECTS

• Understand how to create a class in Python
• Instantiate a class
• Use objects
• Create member functions
• Differentiate between instance and class variables
• Use constructors and destructors
• Understand the type of constructors
10.1 INTRODUCTION TO CLASSES

Classes are real or virtual entities which have an importance to the problem at
hand and sharp physical boundaries. The concept of classes has been discussed
in the previous chapter. This chapter takes these discussions forward. It is easier
to make a class in Python than in any other programming languages. A class in
Python can hold any kind and any amount of data. Those with a C++
background might find the syntax and use of variables odd. As a matter of fact
the mechanism of classes in Python is inspired not just by C++ but also by
Modula-3.
A class in Python can be sub-classed. All types of inheritance, including

multiple inheritances, are supported in Python. Method overriding is also
allowed in Python. The dynamic nature of classes makes Python stand apart
from other languages. Classes can be created at runtime and can even be
changed as and when the program runs.
In a class all data members are public in nature; that is they can be accessed
anywhere in the program. The member functions in a class are all virtual. In a
class all the member functions must have the first argument as the object
representing that class, from now on referred to as 'self'. Interestingly, all the
built in types are themselves classes in Python and they can be extended by the
programmer.
The reader is advised to revisit the chapter on lists. Note that multiple names can
be associated with the same object. Using pointers, for example, an object can be
passed to a function using just one argument and in addition to that, the change
done by the function is visible to the calling function also. In the case of Python,
aliasing (having multiple names for the same object) can be used to accomplish
the above task.
This chapter has been organized as follows. Section 10.2 presents the definition
of a class. Section 10.3 presents the concept of objects and discusses the
instantiation of a class. Section 10.4 discusses the scope of data members.
Section 10.6 discusses constructor overloading. Section 10.7 discusses
destructors and the last section concludes the chapter.
10.2 DEFINING A CLASS

In Python, a class can be defined using the class keyword. The class keyword
is followed by the name of the class. The body of the class follows. It must have
proper indentations.
Syntax
For example the employee class having data members name and age and member
functions getdata() and putdata() can be defined as follows (testing). It was
stated earlier that every function in the class must have at least one argument,
self. The functions of this class have been defined in the traditional way. The
getdata() function here asks for the values of name and age from the user. The
data members are accessed via the self object as they belong to the class and
not just the function. Likewise the putdata() function displays the values of the
data members. Note that the members of a class are accessed via self.
Tip
A class definition has functions but can also have other members.
The attribute of an object is data attribute, and the function that belongs to
an object is method.
10.3 CREATING AN OBJECT

An object is created by associating a name with an instance of the class,
initialized using the default constructor. For example, in creating an object of the
employee class the following statements are used.
e1=employee()
Here, e1 is the name of the object and employee() is the constructor of the class.
An object can also be created using a parameterized constructor, as explained in
the following sections. The creation of an object is referred to as instantiation.
The function of a class can be called using the dot operator with a given class.
For example, to call the getdata() function of the employee class the following
statements are used.
e1.getdata()
Likewise, the other methods of a class can be called using the dot operator.
Code
Tip
An object support the following operations

Instantiation
Attribute references
10.4 SCOPE OF DATA MEMBERS

The scope of a namespace is the region where it is directly accessible. In fact, in
Python scopes are used dynamically. In determining the scope of a namespace,
the following rules are followed:
First of all, the innermost scope is searched
Then the scope of enclosing functions are searched
Then the global namespaces are searched
Then the built-in names are seen
The nonlocal statements rebind the variables in the global scope. In order to
understand the above concept, consider the following code. The following points
concerning the code are worth noting:
The value of a for all instances of the class is 5, until a function that changes
the value of a is called.
In putdata() a does not exist, a is local to getdata()
b can be accessed in both the functions as b is a data member of the class
(note that every time b is called, 'self.b' is used)
On the basis of the above discussion the reader is expected to decode the
following program.
Code
In the following code, 'a' is common for all the classes. 'b' is a member of the
class. Here,'self.b=b' means the data member 'b' of the class (self.b) is
assigned value 'b', which is the second argument of the function getdata().
'c' is local to getdata(), so 'c' of getdata() is not same as that of putdata().
Definition: Instance variable and class variable
An instance variable is one which is unique to each instance and a class variable
is one which is shared by all instances. For example, in the following code b can
be assigned a different value in each instance but c remains the same.
Code
In addition to the above a global data member can be made outside the class,
which is accessible to all the methods (until the scope of the data member is
changed). In the following code 'a' is common for all instances of the class, 'b'
is the data member of the class and 'c' is a local variable.
Code
10.5 NESTING
The designing of a class requires conceptualization of an entity, which has
attributes and behavior. The object of a class can be made in another class also.
That is, a class can also have the objects of another class as its members. This is
called nesting. Note that the attributes of a class can themselves be entities. For
example, in the following code an instance of the date class is created in the
student class. This makes sense, as student is an entity made up of other
entities (like date).
Code
10.6 CONSTRUCTOR
Note that each time a class is instantiated, a constructor (for example, e1=
employee()) is used. In C++ terminology, a constructor is a function which has
the same name as that of the class and initializes the data members. The above
examples used default constructors, which were not made by the programmer.
One can initialize the objects as per the need, by crafting constructors. The
following discussion focuses on two types of constructors: default and
parameterized. A default constructor does not take any argument (for example
the employee() constructor). In Python, the constructors are called using the
functions having the same name as that of the class. However, they are
implemented by making the _init_() function inside the class.
In the following code, the object e1 behaves as expected. The values entered by
the user in the getdata() function are displayed when putdata() is called. In
the case of e2 the function getdata() is not called, therefore the values assigned
in _init_() are displayed.
Code
A parameterized constructor is one which takes arguments - for example in the

following code, the parameterized constructor which takes two parameters name
and age has been created. In order to assign the values to the object, the
instantiation must be of the form:
e2=employee('Naved', 32)
Note that while defining the parameterized __init__, the first parameter is
always 'self' and the rest of the parameters are the values to be assigned to
different data members of the class. In the case of employee class, three
parameters 'self', 'name', and 'age' are given.
Code
10.7 CONSTRUCTOR OVERLOADING

Having the same name function in a class with a different number of parameters,
or different type of parameters, is called function overloading. In C++, Java, C#,
etc., the constructors can also be overloaded - that is one can have more than one
constructors with each having different parameters. In Python, however, we
cannot have more than one _init_ in a class. For example, if we try executing
the following code an error crops up.
The reason is that is one makes a parameterized _init_. Python looks for the
rest of the parameters in the instantiation.
Code
Having studied the importance and implementation of constructors, let us now
implement a constructor and let’s consider the "movie" class, discussed here.
The following code has a movie class, which contains a getdata() and
putdata() function and init (self) for initializing the variables. Note that the
object 'm' does not call the getdata() function but just putdata(). The values
assigned in the constructor are displayed.
Code
10.8 DESTRUCTORS
A constructor initializes the data members of a class and a destructor frees the
memory. The destructor is created using _del_ and called by writing the
keyword del and the name of the object. The following code exemplifies a
destructor in the employee class described in the previous sections.
Code
The next example is the same as the previous one. However, the following code
also demonstrates the use of _class_. _name_, which displays the name of the
object that calls the function. This is useful as the name of the object whose
destructor (or for that matter any method) is being called.
Code
The reader is advised to go through the references for if name = “_main__”:

The docstring associated with the class can be mentioned in the definition of
the class within three double quotes (“”” ...”””).
The docstring associated with the class can be accessed through doc, as
shown in the following example.
Code
The above chapter discusses what is referred to as an instance method. However
another type of method can be created in a class, which is referred to as a class
method.
10.9 CONCLUSION
The last chapter introduced the concepts of object-oriented programming. This
chapter takes the topic further. The chapter introduces the syntax of a class and
the creation of objects. The concept of constructors, their creation, types and
implementation have also been discussed in the chapter. The chapter also
introduces the idea of destructors. Ample examples have been given in the
chapter, which explain the implementation of the concepts introduced earlier.
The following chapter will introduce the idea of inheritance and polymorphism,
which are essential to object-oriented programming. However to be able to
inherit a class or implement operator overloading, one must be versed with the
creation of a class and its use.
GLOSSARY
Data attribute and method: The attribute of an object is data attribute
and the function that belongs to an object is method.
Instance variable and class variable: An instance variable is one which is
unique to each instance and a class variable is one which is shared by all
instances.
Constructor: A constructor initializes the data members.
A parameterized constructor is one which takes arguments.
POINTS TO REMEMBER
The classes in Python can be sub-classed.
All types of inheritance including multiple inheritances are supported in
Python.
A class can be defined using the class keyword in Python.
An object is created by associating a name with an instance of the class,
initialized using the default constructor.
The function of a class can be called using the dot operator with a given
class.
While defining the parameterized init the first parameter is always 'self',
and the rest of the parameters are the values to be assigned to different data
members of the class.
A constructor initializes the data members of a class and a destructor frees
the memory.
The destructor is created using del and called by writing the keyword del
and the name of the object.
__class__.__name__ displays the name of the object that calls the function.
The docstring associated with the class can be accessed through doc.
EXERCISES

1. A class generally has
(a) Function and data members
(b) Function and lists
(c) Lists and tuples
2. A class can have

(a) Any number of functions
(b) Any type of data members
(c) A variable local to a function
3. 'self' is
(a) Object of the same class

(b) Object of the base class
(c) Object of a predefined class
4. Each function in Python must have at least one parameter, which is

(a) Data
(b) List
(c) self
5. The init function

(a) Initializes the data members
(b) Is compulsory
(c) Must be overloaded
6. The init function in a class
(a) Must be overloaded
(b) Can be overloaded
(c) Cannot be overloaded
7. The docstring of a class can be accessed using

(a) __init
(b) __doc__
(c) __class
8. A global variable
(a) Can be accessed anywhere
(b) Can be accessed only in init__
(c) Both of the above
9. The nonlocal variable

(a) Is generally associated and then used
(b) Must not be associated
(c) Does not exist
10. A variable shared by all the instances of a class is

(a) Class variable
(b) Instance variable
(c) Both
11. A variable unique to an instance is
(a) Instance variable
(b) Class variable
(c) Both
12. Which of the following keywords is used to define a class?

(a) class
(b) def
(c) del
13. Which of the following is used to define a function that acts as a destructor?
(a) del
(b) init
(c) Both
14. Which of the following operations are supported by an object?

(a) Instantiation
(b) Attribute reference
(c) Both
15. Suppose e1 is an object, which of the following codes is used to call del?
(a) del e1
(b) e1. del__
(c) Both
16. If the name of the object is to be displayed in a function of a class, then
which of the following can be used?
(a) __class . name__
(b) object . name__
(c) Both
17. In a class all variables are _ by default?

(a) Public
(b) Private
(c) Cannot say
(d) Depends on the type of variables
18. In Python, which of the following operators is used to access methods?

(a) Dot
(b) Plus
(c) []
19. Can a list of objects be created?

(a) Yes, if the type of variables is public
(b) Yes, in all cases
(c) No, in all cases
(d) Yes, if the type of variables is private
20. Data members of a class

(a) Must be private
(b) Can be private
(c) Must be public
THEORY
1. What is an object? How is an object created in Python?
2. Explain the scope of variables in a class. Give an example of data members

of a class which are shared by all the objects and of those which are unique
to an object.
3. What is a constructor? What are the different types of constructors in

Python?
4. Can we overload a constructor in Python?
5. How can one access the name of the docstring in Python?
6. How can one access the name of an object in Python?
7. What is a destructor? How is a constructor created in Python?
8. Give an example of use of a destructor in Python.
9. Give an example of instantiation of a class.
10. Explain the concept of aliasing in Python.
PROGRAMMING EXERCISES
A start-up employs interns. The following details of interns are stored
first_name
last_name
address
mobile_number
e_mail
1. Create a class called Intern, which stores the above details. Craft two
functions getdata() which asks the user to enter data and putdata() to
display the data.
2. In the above program create init which takes only one parameter (self).
3. In question 1, create init, which takes 6 parameters - the first being “self”
and the rest the values of variables stated in question 1.
4. In the above question, craft a destructor.
5. A library management system is to be created, in which the following details

of a “Book” are to be stored.
Name
Publisher
Year
ISBN
Authors
The authors, above, is a list consisting of all the authors of that book.
Create a class called Book, which stores the above details. Craft two
functions - getdata() which asks the user to enter data and putdata() to
display the data.
6. In the above program, create init which takes only one parameter (self).
7. In question 6, create init, which takes 6 parameters with the first being
“self” and the rest the values of variables stated in question 5.
8. In the above question, craft a destructor and call it.
9. Create a class called complex, having real_part and ima_part as its two
data members and getdata() and putdata() as its member functions.
10. In the above question, craft init and del.
11. Create a function called add, which takes two complex numbers as its
parameters and returns the sum of the two complex numbers.
12. Create a function called sub, which takes two complex numbers as its
parameters and returns the difference of the two complex numbers.
13. Create a function called multiply, which takes two complex numbers as its
parameters and returns the product of the two complex numbers.
14. Create a function called div, which takes two complex numbers as its
parameters and returns the result of division of the two complex numbers.
15. Create a class called date having day, month, and year as its data members
and getdata() and putdata() as its member functions. Instantiate the class,
ask user to enter data and display the data.
CHAPTER
11
INHERITANCE
• Understand the concept and importance of inheritance
• Differentiate between inheritance and composition
• Understand the types of inheritance
• Appreciate the role of ‘self’ in methods
• Understand the search in an inheritance tree
• Understand the concept and importance of super
• Appreciate the need of an abstract class
11.1 INTRODUCTION TO INHERITANCE AND

COMPOSITION
Those of you from a C++ background will have studied the importance of
inheritance and composition. Inheritance was projected as a path breaking
concept, which promised to solve all the problems and bring about a change in
the way programming is done. But you must have understood that those who
make such tall claims generally create more problems than they claim to solve.
Inheritance may also create problems; many more than you can imagine.
Many programmers believe that inheritance is a black hole which somehow

attracts programmers, who falls in the trap of tall claims and end up landing
themselves in a situation which tempts them to use multiple inheritance.
Multiple inheritance is like Voldemort, and the object-oriented programming
environment is Hogwart’s. Therefore, it is better to avoid multiple inheritance as
much as possible.
Object-oriented programming has its charms but also comes with its own
problems, it is like demonetization. So, use inheritance only if required. Also
remember never ever to use multiple inheritance. Remember that anything that
can be done using inheritance can be done also in another way too. Composition,
introduced later in the chapter, can be easily used to accomplish most of the
tasks that can be done using inheritance.
In hindsight, inheritance means a class would get features (all or some) from the
parent class. So, when one writes
class SoftwareDeveloper(Employee):
...
it implies that the class SoftwareDeveloper is a subclass of the class Employee.

This relationship falls in the category of an “is a” type relationship. That is,
'SoftwareDeveloper' is-a 'Employee'.
The class from which class(es) are derived from is called a base class and those
that inherit features from the base class are derived classes. In the above
example, Employee is the base class and SoftwareEmpolyee is the derived class.
Note that inheritance does not affect the base class. The derived class can use the
modules of the base class in a variety of ways which are discussed as follows.
11.1.1 Inheritance and Methods

As far as modules are concerned, inheritance can help the programmer to derive
the features by one of the following ways.
The method is not present in the child class, but only in the parent class: In
such cases if an instance of the child class calls the said method, the parent
class’s method is called. For instance, in the following snippet the derived class
does not have a method called show() so calling show using an instance of the
derived class (consider, in this case) calls the method of the parent class.
Code
The method is present in both the parent class and in the derived class: In
such cases, if this method is invoked using an instance of the derived class then
the method of the derived class is called. If the method is called using an
instance of the base class, the method of the base class is called. Note that in
such cases, the derived class redefines the method. This overriding ensures that
the search of the method in the inheritance tree ends up invoking this method
only. For example, in the following snippet x.show() calls the show() method of
the derived class, whereas y.show() calls the method of the base class.
Code
The inherited class modifies the method of the base class and in this process
invokes the method of the base class inside the method of the derived class also.
Note that in the following snippet the show method of the derived class prints a
message, then calls the method of the base class and finally prints another
message. Note that in this case, the method of the base class can be called by
qualifying the name of the method with the name of the base class. For example
in the following snippet the show method of the base class can be called using
ABC.show(self). The importance of the self argument has been explained in
Section 11.3.
Code
The first type of inheritance will henceforth be referred to as implicit

inheritance. In this type the method of the base class can be called using an
instance of the derived class.
The second type of inheritance will henceforth be referred to as explicit

overriding. As stated earlier, the derived class will redefine the method of the
base class and calling this method using an instance of the derived class will
invoke the method of the derived class.
The third type of inheritance is the most important and practical form of
overriding methods. This type of inheritance leaves the room of not making an
instance of the base class, if not required, still using the function.
The following illustration combines the three types of inheritance:
Illustration 11.1: Create a class called Student having _init_ and show
methods. The Student class should have a data member called name. The _init_
should assign value to name and show should display the value. Create another
class called RegularStudent, which will be the derived class of the Student
class. The class should have two methods _init_ and show. The _init_ should
assign values to age and should call the _init_ of the base class and pass the
value of name to the base class. The show method must display the data of the
RegularStudent. In addition to the above both classes should have methods
called random, both of which should be independent of each other (Figure 11.1).
Find what happens when the methods of the base class and the derived classes
are called using the instances of the base and the derived classes.
FIGURE 11.1 Class hierarchy for Illustration 11.1
Solution:
Code
11.1.2 Composition
Making an instance of another class inside a class makes things easy and helps
the programmer to accomplish many tasks. In order to understand the concept,
let us consider an example. Consider that a Student and his PhDguide are
subclasses of the person class. Also, the data of the PhD guide includes the list
of students guided by him/her. This is where composition comes into play. The
instantiation of the students in the PhDGuide class can be done as explained in
the following illustration.
Illustration 11.2: Create a class called Student, having name and email as its
data members and _init_(self, name, email) and putdata(self) as bound
methods. The _init_ function should assign the values passed as parameters to
the requisite variables. The putdata function should display the data of the
student. Create another class called PhDguide having name, email, and students
as its data members. Here, the students variable is the list of students under the
guide. The PhDguide class should have four bound methods: _init_, putdata,
add, and remove. The _init_ method should initialize the variables, the putdata
should show the data of the guide, include the list of students, the add method
should add a student to the list of students of the guide and the remove function
should remove the student (if the student exists in the list of students of that
guide) from the list of students.
Solution:
The details of the classes have been shown in Figure 11.2. It may be noted that
since students is a list therefore a for loop is needed to display the list of
students. Also, while adding the student to the list the data of the passed
parameter has been stored in s (an instance of Student) and s has been added to
the list of the students. The same procedure has been adopted to remove a
student. The code is as follows:
FIGURE 11.2 Details of classes for Illustration 11.2
Code
Output
11.2 INHERITANCE: IMPORTANCE AND TYPES
The concept of classes was introduced in the previous chapter. It was mentioned
that classes are real or conceptual entities which have sharp physical boundaries
and relevance to the problem at hand. A class has attributes (data members) and
behavior (class methods). However, at times these classes must be extended to
be able to solve some specific problem without having to meddle with the
original class. To be able to do so, the language should support inheritance. As a
matter of fact, the presence of classes in the language is primarily because it can
be inherited. Inheritance is, as per most of the authors, one of the most essential
features of object-oriented language.
Using inheritance one can create new classes (derived classes) from an existing
class (base class(es)). Note that a derived class can have even more than one
base class, referred to as multiple inheritance, which is one of the most
undesirable forms of inheritance. Also a base class can itself be a derived class
of some other class. The derived class will have all the allowed features of the
base class plus some features of its own.
A class can be depicted using a class diagram. A class diagram is the

diagrammatic representation of a class, which generally has three sections. In the
representation used next, the first section has the name of the class. The second
section has the attributes and the third section has the class methods. The
following figure (Figure 11.3) shows the class diagram in which the Book class is
the base class and the Text_Book class is the derived class. Note that the arrow is
from the derived class to the base class. The arrow indicates “is derived from” or
“is inherited from.” The next figure (Figure 11.4) gives the details of the two
classes. Note that the Book class has the following attributes:
name
authors
publisher
ISBN
year
The class methods of this class are getdata() and putdata(). The Text_Book
class has another attribute, course. Figure 11.5 shows the class browser showing
the two classes and the relation between them. The corresponding program is
presented in Illustration 11.3.
FIGURE 11.3 Text_book is the derived class of the Book class
FIGURE 11.4 A class diagram generally has three components: the name of the class, the data
members, and the methods of the class. The book class and the textbook class have attributes
and methods as shown in the figure.
FIGURE 11.5 The book examples’ class hierarchy in the class browser of Python
11.2.1 Need for Inheritance

In really large programs, it is difficult to code and debug a class. Once the
programmer has crafted a class, there is little need to meddle with it. If one
needs to craft classes having the same features as the class that has been
developed (and add some more features to it), then it makes sense to derive
classes from the existing class. So, inheritance helps to reuse a code. Reusing the
code has its own advantages. It not only saves time but also money. The
reliability of the program also increases by reusing a code. One can also develop
his class by extending classes developed by others. That is, inheritance helps in
distributing libraries. Inheritance also helps to implement a design that is more
intuitive, better, and more practical. Inheritance also has some disadvantages,
which were discussed in the previous section.
Inheritance is important because of the following factors:

Reusability
Increased reliability
Distributing libraries
Intuitive, better programs
11.2.2 Types of Inheritance

This section presents various types of inheritance and corresponding examples.
Note that the reader is expected to execute the problem given in the illustrations
and analyze the output. As explained earlier, inheritance means deriving new
classes from the existing classes. The classes from which features have been
derived are called the base classes and the class which derives features is called
the derived class. There are five types of inheritance: simple, hierarchical,
multilevel, multiple, and hybrid.
11.2.2.1 Simple Inheritance
The simple inheritance has a single base class and a single derived class.
Illustration 11.3 exemplifies this type. The following illustration has two classes:
Book and Text_Book. The Book class has two methods: getdata and putdata.
The getdata method asks the user to enter the name of the book, number of
authors, the list of authors, publisher, ISBN, and year. The derived class
Text_Book has another attribute called course. The getdata and the putdata
methods extend the base class methods (refer to the previous section).
Illustration 11.3: Implement the following hierarchy (Figure 11.6). The Book
function has name, n (number of authors), authors (list of authors), publisher,
ISBN, and year as its data members and the derived class has course as its data
member. The derived class method overrides (extends) the methods of the base
class.
FIGURE 11.6 The class hierarchy for Illustration 11.3
Solution:
The following code implements the above hierarchy. The output of the program
follows.
Code
Output
11.2.2.2 Hierarchical Inheritance
In hierarchical inheritance, a single base class has at least two derived classes.
Illustration 11.4 exemplifies this type. The following illustration has three
classes: Staff, Teaching, and NonTeaching. Both Teaching and NonTeaching
are the derived classes of the Staff class. The Staff class has two methods:
getdata and putdata. The getdata method asks the user to enter the name and
the salary of the member of the staff. The derived class Teaching has another
attribute called subject. The getdata and the putdata methods extend the base
class methods. Similarly, the derived class NonTeaching has an attribute called
department. The getdata and the putdata methods extend the base class
methods.
Illustration 11.4: Implement the following hierarchy (Figure 11.7). The Staff
function has name and salary as its data members, the derived class Teaching
has subject as its data member and the class NonTeaching has department as its
data member. The derived class method overrides (extends) the methods of the
base class.
Solution:
follows.
Code
Output
11.2.2.3 Multilevel Inheritance
In multilevel inheritance a base class has derived classes which themselves

becomes a base class for some other class. Illustration 11.5 exemplifies this type.
The following illustration has three classes: Person, Employee, and Programmer.
The Person class is the base class. The Employee class has been derived from the
Person class. The programmer class has been derived from the Employee class.
The Person class has two attributes - name and age and two methods - getdata
and putdata. The getdata method asks the user to enter the name and the age of
the member of the staff. The derived class Employee has another attribute
called emp_code. The getdata and the putdata methods extend the base class
methods. Similarly, the class Programmer has another attribute called language.
The getdata and the putdata methods extend its base class methods (Employee
class).
Illustration 11.5: Implement the following hierarchy (Figure 11.8). The Staff
class has name and salary as its data members, the derived class Teaching has
subject as its data member and the class NonTeaching has department as its
data member. The derived class method overrides (extends) the methods of the
base class.
Solution:
follows.
Code
Output
11.2.2.4 Multiple Inheritance and Hybrid Inheritance
In multiple inheritance a class can be derived from more than one base class.
This type of inheritance can be problematic as it can lead to ambiguity. It is
therefore advised to avoid this kind of inheritance as far as possible. However,
the following sections throw some light on this type and the problems associated
with this type.
A design may have a combination of more than one type of inheritance. In the
following figure (Figure 11.9) two classes B and C have been derived from class
A. However, for the class D, the classes B and C are the base classes. This is an
example of combining hierarchical and multiple inheritance. Such a type is
referred to as hybrid inheritance.
FIGURE 11.9 Classes B and C have been derived from A (hierarchical inheritance) and D is
derived from B and C (multiple inheritance)
FIGURE 11.10 Classes B and C have been derived from A (hierarchical inheritance) and D is
derived from B and C (multiple inheritance)
11.3 METHODS
The importance of functions and methods has already been stated in the first
section of this book. Methods are, as stated earlier, just functions with a special
positional parameter within a class. Methods, in fact, help the programmer to
accomplish many tasks. Methods can be bound or unbound. The unbound
methods do not have 'self' as a parameter. While calling such methods, the
first argument must be the instance of the class itself. It is worth mentioning here
that, in Python 3.X, the unbound methods are same as functions whereas in
Python 2.X they are a different type. The bound methods, on the other hand,
have 'self' as the first positional parameter when a method is accessed through
qualifying an instance of a class. Here, the instance of the class needs not to be
passed.
In spite of the above differences, the following similarities between the two
types may not be missed:
A method in Python is also an object. Both bound and unbound methods are
objects.
The same method can be invoked as a bound method and an unbound
method. The discussion and illustrations that follow will clarify the second
point.
11.3.1 Bound Methods

A method can be invoked in a variety of ways. If the first positional parameter of
the method is "self", it is bound. In such cases, the instance of the class can call
the method by passing the requisite parameters.
A variable which holds <Object name>.<method name> (Hari. display), in the

following example, can also be used to invoke the method. Those of you from a
C# background may find the concept similar to that of delegates.
A method can also be invoked by creating an unmanned instance of the class.

The third call of the display method depicts this way of calling method.
Illustration 11.6: Calling a bound method
This illustration has a class called student. The Student class has a display
method, which takes two arguments - the first being the positional parameter and
the second being a string that is printed. Note that the display method is a bound
method and hence is called through an instance of the class.
Code
Output
>>>
========= RUN C:\Python\Inheritance\BoundUnbound.py
=========
Hi I am Hari
Hi I am through X
Caling diaplay again
>>>
11.3.2 Unbound Method

An unbound method does not have self. Therefore the positional parameter
needs not to be passed in method. In such methods, the variables should not be
qualified by 'self'. Calling such methods in the same way as before would
result in an error, as shown in the output of Listing 1 of Illustration 11.7. The
second listing calls the unbound method in an appropriate way. Such methods
must be called by the name of the class and not the object. In Python 3.X, as
stated earlier, such methods work in the same manner as functions. Also note
that normal functions can be called using the class, of which they are members,
as shown in the previous illustration.
Illustration 11.7: Calling an unbound method
This illustration extends the previous illustration and adds the getdata method,
which does not take self as a parameter and therefore is called by the class itself.
Note that this is similar to the static methods in C++.
Code
Output
>>>
========= RUN C:/Python/Inheritance/BoundUnbound.py
=========
Enter the name of the student :Naved
Enter the age of the student :22
File "C:/Python/Inheritance/BoundUnbound.py", line 21, in
<module>
Naved.getdata(name,age)
>>>
Snippet 2:
Code
Output
>>>
======== RUN C:/Python/Inheritance/BoundUnbound1.py ========
Enter the name of the student :Naved
Enter the age of the student :22
Name : Naved
Age : 22
>>>
11.3.3 Methods are Callable Objects

Methods, like any other object in Python, can be stored in a list and called as per
the requirement. In the illustration that follows the class operations has a
constructor _init_(self, number), which assigns the value of the second
parameter to the data member called number. The class has two methods square
and cube. The first method calculates (and returns) the square of the number and
the second calculates (and returns) the cube of the number. Two instances of the
class operations have been created: X and Y. X is initialized to 5 and Y to 4. The
list List stores the objects X.square, X.cube, Y.square, and Y.cube. The
elements of the list are then called one by one and invoked.
Illustration 11.8: Methods as callable objects
Code
Output
>>>
======= RUN C:/Python/Inheritance/CallableObjects.py =======
25
125
16
64
>>>
11.3.4 The Importance and Usage of Super

A class may have data members and member functions (method). A method is
just a function in a class, defined using the keyword 'def'. As discussed in the
earlier chapters, the methods depict the behavior of a class. Generally, the
method’s first argument is an instance of the class itself. The first argument,
generally referred to as self, is similar to 'this' of C++. Using self with the
variable name indicates that the reference is to the instance variable, not that in
the global scope. For example in the following snippet, the _init_ method has
two arguments: the first being self and the second being the name. Assigning
name to self.name implies that the local variable name is given value “name”
(the second argument of _init_). Similarly, the putdata method has a positional
parameter indicating the data for the instance which invokes putdata must be
shown. Note that the output reinforces the fact that self-binds the method call
with the instances.
Code
Output
============ RUN C:/Python/Inheritance/Basic.py ============
name : Hari
name : Nakul
>>>
However, methods can also be crafted without the 'self' argument. These are
unbound methods. The concept has been discussed in Section 11.3.2 of this
chapter. A method of a class is an instance method by default. So, generally, the
method of a class can be called by creating an instance of the class and using the
dot operator to call the method. Note that this was the case in languages like C#,
Java, etc. Note that there are other ways of invoking a method, as discussed in
this chapter.
However, there are other types of methods as well. For example, the static
methods do not require the instance of a class as their first argument.
11.3.5 Calling the Base Class Function Using Super

The functions of the base class can be called using super. In fact, super can be
used to call any function of the base class and it clearly depicts the calling of the
base class’s function. In order to understand the usage of super, let us go through
the following example. In the following example, BaseClass has two methods:
_init_ and printData. _init_ has one positional parameter and one parameter
that initializes “data” (the data member of the BaseClass.) The DerivedClass is
the derived class of the BaseClass. This class has _init_, which takes a
positional parameter (self), and two other parameters. The first initializes the
data member of DerivedClass and the second is passed onto the init of the base
class (BaseClass) using super. The super takes the name of the class
(DerivedClass), the positional parameter (self) and calls the _init_ of the
base class by passing all parameters except the positional parameter. Note that
the second function also uses super in the same manner.
Code
Output
>>>
========== RUN C:/Python/Inheritance/superDemo.py ==========
Data of the base class : 4
Data of the base class : 5
Data of the derived class : 6
>>>
11.4 SEARCH IN INHERITANCE TREE

An object is searched for in the inheritance tree in a bottom up fashion. First of
all, the class is searched for the given object. If it is found then the found object
is used to accomplish the given task. If not, then its super class (base class) is
searched for the object. In case of more than one base class, ambiguity can
occur.
For example, in the following illustration the Derived1 class has been derived
from BaseClass. The show() method of this class displays the values of 'data1'
and 'data'. The former is in the class and therefore its value is displayed.
However the former is not in the class, so the inheritance tree is searched for the
object. Note that 'data' exists in the base class (BaseClass) and therefore its
value will be displayed. This is true for methods also. Even if the derived class
does not have a particular method, it can be invoked if it exists in the parent
class or in any other class up the inheritance tree. It may also be stated here that
the objects are generally searched from left to right in a particular level.
FIGURE 11.11 The class hierarchy for given illustration
Code
Output
======= RUN C:/Python/Inheritance/InheritanceTree.py =======
Data : 1
1
Data : 3
Base class data : 2
Data : 5
Base class data : 4
>>>
11.5 CLASS INTERFACE AND ABSTRACT CLASSES

At times the classes are crafted so that they can be sub classed. While
designating, there is no intention of instantiating the class. The classes which
will not be instantiated but will only be used to create base classes are called
abstract classes. In order to understand the concept, let us consider an example.
The following example has four classes: BaseClass, Derived1, Derived2, and
Derived3.
The BaseClass has two methods: method1 and method2. The first method has
some task associated with it, whereas the second wants the derived class to
implement it. The derived class should, to be able to call this method, have a
method called action. The first derived class (Derived1), replaces method1. So
if an instance of Derived1 calls method1, the version defined in Derived1 would
be called. The second method extends method1, it adds something to method1
and also calls the BaseClass version of method1. When method1 is called from
Derived2, the BaseClass version is called as the search in the inheritance tree
invokes the base class version of the method. The third derived class (Derived3)
also implements the action method defined in the base class. Note that when
method2 is called through an instance of Derived3, the base class version of
method2 is invoked. This version calls action and a new search is initiated, thus
resulting in the invocation of action of Derived3. Illustration 11.9 presents the
code.
Note that the above concept can be extended and a class may have methods that
would be implemented by the derived classes. Interestingly, Python has
provisions that such classes would not be instantiated until all such methods are
not defined. Such base classes are called abstract classes. The implementation of
abstract classes has been discussed in the appendix of this book.
Illustration 11.9: Implement the following hierarchy. 'method1' of Derived1

should replace method1 of the base class, method1 of Derived2 should extend
method1 of the base class and action of Derived3 should implement method2 of
the BaseClass.
FIGURE 11.12 Class hierarchy for Illustration 11.9
Solution:
Output
>>>
============ RUN C:/Windows/System32/Abstract.py
============
Class : <class '_main__.Derived1'>
A new method, has got nothing to do with that of the base
class
A method that extends the base class method
In BaseClass from method1
In BaseClass from method1
Implementing the base class method
FIGURE 11.13 Searching in an inheritance tree
11.6 CONCLUSION
The chapter introduces the concept of inheritance, which is one of the most
important ingredients of object-oriented programming. Inheritance, as explained
in the chapter, helps the program in reusing the code and making the program
more structured. However it should be used wisely, as in many cases it leads to
problems like ambiguity. The reader must also appreciate that it is not always
necessary to use inheritance. Most of the tasks can be accomplished using
composition. However even if using inheritance becomes a necessity, be clear
about the type of inheritance required, the type of method calls required and the
use of bound methods. The discussion on object-oriented programming
paradigms continues in the next chapter also, where the concept of operator
overloading has been introduced. The last chapter, this chapter and the next one
will help the reader to successfully develop a software using OOP.
GLOSSARY
Inheritance: Inheritance is the process of creating subclasses from existing
classes.
Base class and derived class: The class from which other classes are derived is
the base class and the classes that inherit from the base class are the derived
classes.
Implicit inheritance: In this type, the method of the base class can be called
using an instance of the derived class.
Explicit overriding: In this type of inheritance, the derived class will redefine
the method of the base class and calling this method using an instance of the
derived class would invoke the method of the derived class.
POINTS TO REMEMBER
Inheritance provides reusability and increased reliability.
Types of inheritance are simple inheritance, multiple inheritance, multilevel
inheritance, hierarchical inheritance, and hybrid inheritance.
Multiple inheritance may lead to ambiguity.
A bound method has a “self” parameter whereas an unbound method does
not have “self” parameter.
A class can also be instantiated in another class.
Super can be used to access the base class methods.
The inheritance tree is searched to find the version of method to be invoked.
EXERCISES
1. A class that cannot be instantiated until all its methods have been defined by
its subclass(es) is called
(a) Abstract class
(b) Meta class
(c) Base class
A class called operation has an_init__, which takes a positional parameter
and an integer as an argument. Two instances of operations Num1 and Num2
have been defined as follows. The class has two functions; the first
calculates the square of a number and the second calculates the cube. A list
called List1 is created which contains the names of the four methods (two
of Num1 and two of Num2). A for loop is then used to call the methods as
shown in the following snippet:
Num1=operations(5)
Num2=operations(4)
List= [Num1.square, Num1.cube, Num2.square, Num2.cube]
for callable_object in List:
print(callable_object())
2. The program containing the above code (assume that the rest of the code is
correct)
(a) Has no syntax error but does not execute
(b) Has syntax error
(c) Has no syntax error and executes
(d) Insufficient information
3. In the above question, what would the output be (if the code is correct)?
(a) The code is not correct
(b) 25 125 16 64
(c) 125 25 64 16
4. The names of the methods in the list (question 2) are similar to (in ‘C#’)
(a) Meta classes
(b) Delegates
(c) Both
5. 'self' in Python is similar to
(a) 'this' is ‘C#’
(b) 'me' in ‘C#’
(c) Delegate in ‘C#’

6. class Student:
def display(self, something):
print("\n"+something)
def getdata(name,age):
name=name
age=age
print("\nName\t:",name,"\nAge\t:",age)
In the above snippet, how would you invoke getdata? (assume that Hari is
an instance of Student).
(a) Student.getdata("Harsh", 22)
(b) Hari.getdata("Harsh", 24)
(c) Both are correct

7. Can a method also be invoked by creating an un-named instance of a class?

(a) Yes
(b) No
(c) Insufficient data
(d) There in nothing called unnamed instance of a class in Python
8. Which of the following is used in searching an inheritance tree?

(a) Breadth First search
(b) Depth First search
(c) Both
9. In an inheritance tree at the same level, which policy is used to search an

object?
(a) Left to right
(b) Right to left
(c) Any
10. 'super' can be used to call
(a) The init of the base class

(b) Any method of the base class
(c) Cannot be used to call methods of the base class
(d) None of the above. The usage of 'super' depends on the type of
inheritance
11. Which type of inheritance leads to ambiguity?

(a) Multiple
(b) Multilevel
(c) Both
(d) None
12. Which type of inheritance has just one base class and a single derived class?
(a) Simple
(b) Hierarchical
(c) Multiple
13. Which type of inheritance has more than one base class(es) and a single
derived class?
(a) Simple
(b) Hierarchical
(c) Multiple
14. Which type of inheritance has more than one derived class(es) and a single
derived class?
(a) Simple
(b) Hierarchical
(c) Multiple
15. Can a derived class be a base class of some other class?

(a) Yes
(b) No
(c) Insufficient data
THEORY
1. What is inheritance? Explain the importance of inheritance.
2. What are the disadvantages of inheritance? Explain with reference to

multiple inheritance.
3. What are the various types of inheritance? Give examples.
4. What are the problems in implementing multiple inheritance? How are they
resolved?
5. What is composition? Is it a type of inheritance?
6. Is inheritance mandatory in object-oriented programming? Justify.
7. What is the difference between “is a” and “has a” relationships? Explain

with the help of examples.
8. Which is better, inheritance or composition? Can all that can be achieved
using inheritance be done using composition?
9. Explain the use of “super.” How can it be used to call methods of the base
class?
10. Are methods in Python objects? Justify your answer. What is meant by
callable object?
11. What is an abstract class? How does an abstract class help in achieving the
goals of OOPs?
12. What are bound methods? What are the various ways of invoking a bound
method?
13. Differentiate between a bound and an unbound method. Give examples.
14. What is the importance of 'self' in Python?
15. Explain the mechanism of search in an inheritance tree.
1. A class called Base1 has two methods: method1(self, message) and
method2(self). The first method prints the message passed as an argument
to the method. The second invokes another method called action1(self),
which would be defined by the sub-class (Derived2) of Base1. Derived1,
another derived class of Base1, redefines method1 and does nothing with
method2. Derived2, on the other hand, does nothing with method1.
Implement the hierarchy and find what happens in the following cases:
(a) An instance of Base1 calls method1
(b) An instance of Derived1 calls method1
(c) An instance of Derived2 calls method1
(d) An instance of Base1 calls method2
(e) An instance of Derived2 calls method2
(f) An instance of Derived1 calls method2
(g) An instance of Derived2 calls action
2. A class called operation has an _init_, which takes a positional parameter

and an integer as an argument. The class has two functions; the first
calculates the square root of a number and the second calculates the cube
root. Two instances of operations, Num1 and Num2 are to be created. A list
called List1 is to be created which contains the names of the four methods
(two of Num1 and two of Num2). Implement the above and use a for loop to
call all the callable objects from the list.
3. A class employee has two methods getdata(name, age) and

getdata1(self, name, age). The getdata method stores the values in the
local variables. Another method called putdata shows the data. Write a
program to call the methods (the first is not bound but the second is) and
display the data.
4. Create the following hierarchy and explain the search process of method
called "show".
CHAPTER
12
OPERATOR OVERLOADING
• Understand the importance of operator overloading
• Implement constructor overloading
• Understand and be able to use various methods used for overloading operators
• Implement operator overloading for complex numbers and fractions
• Understand the importance of destructors
12.1 INTRODUCTION
Operators are defined for primary data structures in all the languages. For
example, in Python the + operator adds two numbers or two floats or
concatenates two strings. However, for user defined data types the programmer
can't use these operators directly.
Operator overloading helps the programmer to define existing operators for user
defined objects. This makes the language powerful and the work simple. This
simplicity and intuitiveness in turn makes programming fun.
In Python, operator overloading is done by defining specific methods discussed

in the sections that follow. Operator overloading can be used to intercept Python
operators by classes and even to overload built-in operations, e.g. attribute
access. The said methods which help in operator overloading have been specially
named and Python calls these methods when instances of classes use the
associated operator. Moreover, it’s not always the case that operator overloading
must be implemented.
This chapter discusses operator overloading. The chapter is organized as follows.

Section 12.2 revisits _init_ and discusses how it can be overloaded. Section
12.3 presents some of the common operator overloading methods. Section 12.4
presents an example of overloading the binary operators. Section 12.5 discusses
_iadd_. Section 12.6 discusses the comparison operators. Section 12.7 discusses
bool and len and the next section presents the concept of destructors. The last
section concludes the chapter.
12.2 _INIT_ REVISITED
The _init_ function has already been explained in Chapter 10 of this book. The
function initializes the members of a class. Those of you from a C++ or Java
background will find it hard to ignore the similarity between the constructors
(which have the same name as that of the class in C++, etc.) and the _init_
function. Earlier it was stated that _init_ cannot be overloaded, which is partly
true. Though one cannot have two _init_ functions in the same class, there is a
way to implement constructor overloading as explained in the following
discussion.
As stated earlier the purpose of _init_ is to initialize the members of the class.
In the following example (Illustration 12.1), a class called complex has two
members: real and imaginary, which are initialized by the parameters of the
_init_ function. Note that the members of the class are denoted by self.real
and self.imaginary and the parameters of the functions are initialized by real
and imaginary. The example has a function called putData to display the values
of the members. In the _main_() function, c1 is an instance of the class complex.
The object c1 is initialized by 5 and 3 and the putData() of the class has been
invoked to display the 'Complex Number'.
Illustration 12.1: Create a class called complex, having two members - real and
imaginary. The class should have _init_, which takes two parameters to
initialize the values of real and imaginary respectively and a function called
putData to display the complex number. Create an instance of the complex
number in the _main_() function, initialize it by (5, 3) and display the number
by invoking the putData function.
Code
Output
=== RUN C:\Python\Operator Overlaoding\_init_\Example1.py
===
5 + i 3
>>>
Let us consider another example of _init_. The example (Illustration 12.2) deals
with the implementation of vectors. In the example, a class called Vector has
two data members called args and length. Since arg can contain any number of
items, the _init_ has *args as the parameter. The putData function displays the
vector and the _len_ function calculates the length of the Vector (as in the
number of arguments).
Illustration 12.2: Create a class called Vector, which can be instantiated with a
vector of any length. Design the requisite _init_ function and a function to
overload the len operator. The class should also have a putData function to
display the vector. Instantiate the class with a vector having:
no element
one element
two elements
three elements
Display each vector and also display the length.
Code
Output
>>>
=== RUN C:\Python\Operator Overlaoding\_init_\Example2.py
===
()
Length 0
(2,)
Length 1
(3, 4)
Length 2
(7, 8, 9)
Length 3
>>>
Note that in the above example, _init_ has the same effect as having many
constructors with different parameters. Although _init_ has not been
overloaded in the literal sense, the program has the same effect as that of one
having overloaded constructors.
12.2.1 Overloading _init_ (sort of)

Constructors can be overloaded by assigning None to the arguments (some or all,
except for the positional argument). In order to understand the point, consider a
class called Complex. The class must have two constructors; one which takes two
arguments and one when no argument is given. In the first case the real and
imaginary part of Complex should be initialized with the arguments of _init_
and in the second the real and imaginary should become zero. One of the
simplest solutions is to check if the two arguments are None or not. If both of
them are None, the data members should be made zero. In the second case they
should contain the arguments, passed in _init_. Though the following program
(Illustration 12.3) does not have two _init_, nevertheless the above task has
been accomplished.
Illustration 12.3: Construct a class called Complex having real and ima as its
data members. The class should have an _init_ for initializing the data members
and putData for displaying the complex number.
Code
Output
>>>
== RUN C:/Python/Operator Overlaoding/_init_/Example 10.py
==
5 +i 3
0 +i 0
>>>
12.3 METHODS FOR OVERLOADING BINARY OPERATORS
The following methods (Table 12.1) help in overloading the binary operators like
+, –, *, and /. The operators operate on two operands: self and another instance
of the requisite class. When an operator is used between objects, the
corresponding methods are invoked. For example for objects c1 and c2 of a class
called Complex, c1+c2 invokes the _add_ method. Similarly, the – operator
would invoke the _sub_ method, the * would invoke the _mul_ method and so
one. Table 12.1 shows the method and the operator due to which the method is
invoked.
Table 12.1 Methods for overloading binary operators
Task Method Explanation

Addition _add_ Helps in overloading the + operator. Generally,
this takes two arguments: the positional
parameter and the instance to be added.
Subtraction _sub_ Helps in overloading the – operator. Generally,
parameter and the instance to be subtracted.
Multiplication _mul_ Helps in overloading the * operator. Generally,
parameter and the instance to be multiplied.
Division _truediv_ Helps in overloading the / operator. Generally,
parameter and the instance to be divided.
The use of the above operators has been explained in the following section.
12.4 OVERLOADING BINARY OPERATORS: THE

FRACTION EXAMPLE
The overloading of the operators shown in the above table can be easily
understood by the example that follows. The following example overloads the
addition (+), subtraction (-), multiplication (*), and division (/) operators for a
class fraction. The fraction class depicts the standard fraction, having a
numerator and a denominator. The details of the methods of the class are as
follows:
1. _init_
The _init_ initializes the class by setting the numerator to 0 and the
denominator to 1. The statement
x=fraction()
therefore, creates a fraction 0/1.
2. _add_
The _add_ helps in overloading the + operator. The statement

z=x+y
calls the _add_ of x and takes y as the “other” argument. Therefore, it must
have two arguments: a positional parameter (self) and a fraction (other).
The addition of two fractions is done as follows. The LCM of b1
and b2 becomes the denominator of the resultant fraction. The numerator is
calculated using the following formula.
Note that the resultant fraction is stored in another fraction (s). The method
_add_ returns s.
3. _sub_
The _sub_ helps in overloading the - operator. The statement

t=x-y
calls the _sub_ of x and takes y as the “other” argument. Therefore, it must
have two arguments: a positional parameter (self) and a fraction (other).
The difference of two fractions is done as follows. The LCM of b1
and b2 becomes the denominator of the resultant fraction. The numerator is
Note that the resultant fraction is stored in another fraction (d). The method
_sub_ returns d.
4. _mul_
The _mul_ helps in overloading the * operator. The statement

prod=x*y
calls the _mul_ of x and takes y as the “other” argument. Therefore, it must
have two arguments: a positional parameter (self) and a fraction (other). The
product of two fractions is calculated as follows. The numerator is
numerator = a1 × a2
And the denominator is calculated as follows.
denominator = b1 × b2
Note that the resultant fraction is stored in another fraction (m). The method
_mul_ returns m.
5. _truediv_
The _truediv_ helps in overloading the / operator (which returns an

integer). The statement
div = x/y
calls the _truediv_ of x and takes y as the “other” argument. Therefore, it

must have two arguments: a positional parameter (self) and a fraction
(other).
The division of two fractions is done as follows. The numerator is

numerator = a1 × b2
and the denominator is calculated as follows.
denominator = b1 × a2
Note that the resultant fraction is stored in another fraction (answer). The
method _truediv_ returns the answer.
Illustration 12.4: Create a class called fraction having numerator and

denominator as its members. Overload the following operators for the class
+
–
*
/
Create LCM and GCD methods in order to accomplish the above tasks. The
LCM method should find the LCM of two numbers and the GCD method should
find the GCD of the two numbers. Note that LCM(x, y) × GCD(x, y) = x × y.
Solution: The implementation has already been discussed. The following code
performs the requisite tasks and the output follows.
Code
Output
== RUN C:/Python/Operator Overlaoding/Add/Fraction_add.py ==
Enter the numerator :2
Enter the denominator :3
First fraction :
2 / 3
Enter the numerator :4
Enter the denominator :5
Second fraction :
4 / 5
Sum :
22 / 15
Difference :
-2 / 15
Product
8 / 15
Division
10 / 12
>>>
Was it really needed?
Note that the above illustration has been included in the chapter to explain the
overloading of binary operators. Python, as such, provides addition, subtraction,
multiplication and division for fractions (refer to Chapter 2). The same task can
be done without overloading the operators as follows:
>>> from fractions import Fraction
>>> X=Fraction (20,4)
>>> X
Fraction (5, 1)
>>> Y=Fraction (3,5)
>>> Y
Fraction (3, 5)
>>> X+Y
Fraction (28, 5)
>>> X-Y
Fraction (22, 5)
>>> X*Y
Fraction (3, 1)
>>> X/Y
Fraction (25, 3)
>>>
12.5 OVERLOADING THE += OPERATOR

The += operator adds a quantity to the given object. For example if the value of
“a” is 5, then a+=4 would make it 9. However, the operator works for integer,
reals and strings. The use of += for integer, real and string has been shown as
follows:
>>> a=5
>>> a+=4
>>> a
9
>>> a=2.3
>>> a+=1.3
>>> a
3.5999999999999996
>>> a="Hi"
>>> a+=" There"
>>> a
'Hi There'
>>>
However, in order to make it work for a user defined data type (or object), it
needs to be overloaded. The _idd_ helps in accomplishing this task. The
following illustration (Illustration 12.5) depicts the use of _idd_ for an object of
the complex class. A complex number has a real part and an imaginary part.
Adding another complex number to a given complex number adds their
respective real parts and imaginary parts. The program follows. Note that,
_iadd_ takes two arguments. The first being the positional parameter and the
second another object called “other.” The real part of “other” is added to the real
part of the object and the imaginary part of “other” is added to the imaginary
part. The _iadd_ returns “self.” Likewise, the reader may overload the _iadd_
operator for his class, as per the requirements.
Illustration 12.5: Overloading += for complex class (Illustrations 12.1 and 12.3)
Code
Output
12.6 OVERLOADING THE > AND < OPERATORS

The greater than (>) and less than (<) operators work in the usual manner for the
integers, fractions and some other predefined types. However to be able to use
these operators for user defined classes, the programmer must overload the
operators. In Python, greater than (>) and less than (<) can be overloaded using
the _gt_ and _lt_. The _gt_ returns a true or a false depending upon whether
the first object is greater than the second or not. Similarly, the _lt_ returns a true
or a false depending upon whether the first object is less than the second or not.
The following example overloads the _gt_ and _lt_ for a class called Data. The
Data class has a data member called “value.” The _gt_ compares the value of
the instance (self) and that of another instance (other). If the value of the
instance is greater than that of the other instance then a true is retuned, otherwise
a false is returned. Similarly, The _lt_ compares the value of the instance (self)
and that of another instance (other). If the value of the instance is smaller than
that of the other instance then a true is returned, otherwise a false is returned.
Illustration 12.6: Write a program to create a class called Data having “value”
as its data member. Overload the (>) and the (<) operator for the class.
Instantiate the class and compare the objects using _lt_ and _gt_.
Solution: The mechanism of the _gt_ and _lt_ has already been discussed. The
program follows.
Code
Output
>>> `
==== RUN C:/Python/Operator Overlaoding/Comparision.py ====
True
False
>>>
print(X<Y)
12.7 OVERLOADING THE _BOOLEAN_ OPERATORS:

PRECEDENCE OF _BOOL_OVER _LEN_
In using “if” and “while” the programmer checks the condition passed in “if” or
“while.” If the condition is true the block following “if” or “while” is executed,
otherwise it is not executed. We can also define the Boolean operators for a user
defined object. In order to accomplish this, the programmer will require some
method that helps in the overloading. Python provides two Boolean operators -
_bool_ and _len_. The _bool_ method returns a true if the requisite condition is
met, otherwise it returns a false. The _len_ method finds the length of the data
member and returns false if it is null. The Boolean condition can be checked
using the _len_ method, only if the _bool_ for that class is not defined. In case
both _len_ and _bool_ are defined in a class, _bool_ takes precedence over
_len_.
For example in the following illustration, writing if (X), where X is an instance

of the class returns a false if no argument is passed while instantiating the class.
Note that the first listing uses _len_. The next illustration (Illustration 12.7)
checks the length of the data member “value” to return a true if “value” is not
null and false otherwise.
Illustration 12.7: The following illustration creates a class called data. If no

argument is passed while instantiating the class a false is returned, otherwise a
true is returned.
Program
class Data:
def_len_(self): return 0
X=Data()
if X:
print("True returned")
else:
print("False returned")
Output
>>>
======== RUN C:/Python/Operator Overlaoding/len.py ========
False returned
>>>
Illustration 12.8: Another variant of the above example has 'value' as its data
member. If the 'value' is null a false is returned, otherwise a true is returned.
Program
Output
======== RUN C:/Python/Operator Overlaoding/len.py ========
True returned
False returned
>>>
Also note that if _bool_ is also defined in the class, then it takes precedence over
the _len_ method. The _bool_ returns a true or a false as per the given
condition. Although overloading _bool_ may not be of much use as every object
is either true or false in Python. Illustration 12.9 presents an example in which
both _bool_ and _len_ are defined.
Illustration 12.9: An example in which both _bool_ and _len_ are defined.
Program
Output
====== RUN C:/Python/Operator Overlaoding/_bool_.py ======
From Bool
True returned
From Bool
False returned
>>>
12.8 DESTRUCTORS
Destructors are called automatically when the space of an object of a class is
reclaimed. Destructors are complementary to constructors. Constructors are
called when a new object is created and the destructors are called when its space
is reclaimed. It may be noted that the destructors are not as important in Python
as they are in some other object-oriented languages. In fact, many programmers
consider the destructors as “obsolete.” The reasons for this are as follows.
Garbage collection: Garbage collection is one of the characteristics of Python.

That is, the memory allocated to an object is reclaimed as soon as the object (an
instance of a class) is reclaimed. So, there is little need of an explicit destructor.
Moreover, it becomes difficult to predict the position at which the destructor

needs to be called. For example, in the following listing the destructor is called
as soon as a value is assigned to the object. But conceptually, the user may want
to use the object in the remaining task.
No garbage collection: There is another reason for not using destructors. They,
at times, hinder garbage collection. This is good, if deliberate, but bad if not used
without due deliberation. Most of the time the garbage collection done by
Python is good. The following illustration (Illustration 12.10) presents a class
called Data, which has an explicit destructor (_del_). Note that when the
memory is reclaimed, a destructor is called.
Illustration 12.10: Use of destructor
Code
Having discussed the problems of destructor, it is always better to craft an

explicit method for the termination activities rather than allowing a destructor to
be called.
12.9 CONCLUSION
Overloading means many meanings for the same symbol. Primarily overloading
can be segregated into two classes: name overloading and symbol overloading.
An operator, which is a symbol that tells the compiler to perform some
mathematical operation, can also be overloaded. That is, operator overloading
can be defined as giving multiple definitions to an operator. Python allows
operators to be overloaded for user defined data types. In Python, operator
overloading can be accomplished by overriding the requisite methods and calling
them as and when required. It may be stated here that the overload operator is a
member of a class. This chapter introduces the concept of operator overloading
and explains the idea using ample examples. The reader is expected to attempt
the problems given in the exercise to get a clear insight and be able to use the
methods introduced in the chapter in practical situations.
GLOSSARY
Operator overloading: This is the mechanism of assigning a new meaning
to an existing object.
POINTS TO REMEMBER
Operator overloading helps the programmer to define an existing operator
for user defined objects.
In Python all expression operators can be overloaded.
Operator overloading can be implemented using special methods.
_bool_ has higher priority over _len_.
EXERCISES
1. Using operator overloading, the programmer can

(a) Define an existing operator for user defined data type
(b) Create new operators
(c) Both
2. In Python, operator overloading can be implemented by

(a) Defining corresponding methods in the class for which user defined
objects would be made
(b) Operators are redefined in the same way as in C++
(c) Python has predefined methods for defining operators
3. Can _init_ be overloaded?

(a) Yes
(b) No
(c) It can be overloaded only for specific classes
4. If the same _init_ is to be designed to accept varying number of arguments,

which of the following is the correct representation?
(a) def _init_(self)
(b) def _init_(self, *args)
(c) def _init_(self, args)
(d) Both (b) and (c)
5. The above task can be accomplished by

(a) Not giving any arguments in _init_
(b) Equating some of the arguments to NONE
(c) Both
6. Which of the following methods is used to overload the + operator?

(a) _add_
(b) _iadd_
(c) _sum_
7. Which of the following is used to overload the – operator?

(a) _diff_
(b) _sub_
(c) _minus_
8. Which of the following is used to overload the * operator?

(a) _prod_
(b) _mul_
(c) Both
9. For which of the following, operator overloading is really needed?

(a) Complex
(b) Fraction
(c) Polar coordinates
10. Which of the following is overloaded using _iadd_?

(a) +
(b) + =
(c) ++
11. Can > and < operators be overloaded in Python?

(a) Yes
(b) No
(c) Only for specific classes
12. Which has more priority _bool_ or _len_?

(a) _bool_
(b) _len_
(c) Both
THEORY
1. What is operator overloading? Explain its importance.
2. Explain the mechanism of overloading operators in Python.
3. Can all Python operators be overloaded?
4. The membership can be tested using the 'in' operator. The contains
method can be used for testing the membership in Python. Create a class
having three lists and overload the membership operator for the class.
5. Explain the following methods and explain operator overloading using the
operators.
(a) _add_
(b) _iadd_
(c) _sub_
(d) _mul_
(e) _div_
(f) _len_
(g) _bool_
(h) _gt_
(i) _lt_
(j) _del_
6. The following methods have not been discussed in the chapter. Explore the
following and use them for complex class.
(a) _getitem_
(b) _setitem_
(c) _iter_
(d) _next_
PROGRAMMING
1. Create a class called Distance having meter and centimeter as its data
members. The member functions of the class would be putData(), which
takes the values of meter and centimeter from the user; putData(), which
displays the data members and add, which adds the two distances.
The addition of two instances of distances (say d1 and d2) would require
addition of corresponding centimeters (d1.centimeter +s2.centimeter), if the
sum is less than 100, otherwise it would be (d1.centimeter
+s2.centimeter)%100. The “meter” of the sum would be the sum of meters
of the two instances (d1.meter +d2.meter), if (d1.centimeter +d2.centimeter)
<100, otherwise it would be (d1.meter +d2.meter+1).
2. Overload the + operator for the above class. The + operator should carry out
the same task as is done by the add function.
The subtraction of two instances of distances (say d1 and d2) would require
the subtraction of corresponding centimeters (d1.centimeter -s2.centimeter).
The “meter” of the difference would be the sum of meters of the two
instances (d1.meter - d2.meter).
3. Overload the – operator for the distance class. Assume that d1-d2, would
always mean d1>d2.
4. Overload the += operator for Distance class. The += operator (that is

d1+=d2) would require the addition of d1 and d2 (as explained earlier) and
updating d1 with (d1+d2). Note that, the value of d2 is not altered.
5. Overload the * operator for the Distance class.

The government of a developing country intends to do away with the present
currency and intends to introduce a barter system, in which 12 bottles of
“Tanjali” would be equivalent to a unit of currency. This in turn would
increase the sales of the company. Hari and Aslam have 37 and 92 bottles of
‘Tanjali’ and would like to exchange the bottles to buy tickets to a movie. If
each ticket is 60 Units, would they be able to watch the movie?
6. Now, help the people of the country by developing a program having a class
called nat_currency and overload the + operator, which adds two instances
of nat_currency.
7. For the above question, overload the – operator.
8. For the nat_currency class of question 6, overload the += operator.
9. For the nat_currency class of question 6, overload the * operator.
10. Create a class called date having members’ dd, mm, and yyyy (date, month,
and year). Overload the + operator, which adds the two instances of the date
class.
A hypothetical number called irr, of the form c , has b constant. Two

instances of irr can be added as follows. If the first irr number is r1 = a1 +
c1 and the second is r2 = a2 + 2 , the addition of r1 and r2 would be r =
r1 + r2 = (a1 + a2) + (c1 + c2) .
The difference of r1 and r2 would be r = r1 – r2 = (a1 – a2) + (c1 – c2) .
The product of r1 and r2 would be r = r1 × r2 = (a1a2 + c1c2 d) + (a1c2 +
a2c1) .
11. Create a class called irr and overload the + operator.
12. For the irr class, overload the – operator.
13. For the irr class, overload the += operator

14. For the irr class, overload the * operator.
A vector is written as a + b + c , where is a unit vector in the x axis, is
the unit vector in the y axis and is the unit vector is the z axis. The
addition of two vectors requires the addition of the corresponding
components, addition of the corresponding components and the addition of
the corresponding components. That is, for two vectors v1 = a1 + b1 + c1
and v2 = a2 + b2 + c2 , the sum would be v = v1 + v2 = (a1 + a2) + (b1 +
b2) + (c1 + c2) . Likewise, the difference of two vectors requires the
subtraction of the corresponding components, subtraction of the
corresponding components and the subtraction of the corresponding
components. That is, for two vectors v1 = a1 + b1 + c1 and v2 = a2 + b2 +
c2 , the difference would be v = v1 – v2 = (a1 - a2) + (b1 - b2 ) + (c1 - c2 ) .
15. Create a class called vector having three data members a, b, and c. The class
must have the getData() function to ask the user to enter the values of a, b,
and c; the putData() function to display the vector.
16. Overload the + operator for the vector class.
17. Overload the – operator for the vector class.
18. Overload the += operator for the vector class.

CHAPTER
13
EXCEPTION HANDLING
• Understand the concept of exception handling
• Appreciate the importance of exception handling
• Use try/except
• Manually throw exceptions
• Craft a program that raises user defined exceptions
13.1 INTRODUCTION
Writing a program is an involved task. It requires due deliberation, command
over the syntax and problem solving capabilities. In spite of all efforts, there is a
possibility of some error cropping up or of an unexpected output. These errors
can be classified as follows. The first types of errors are those due to syntax or
those which can be intercepted by the compiler. On compiling a program having
such errors, some standard message appears. These can be handled by learning
the syntax or changing the code as per the requirement of the problem at hand.
The following is an example of a code having syntax error. Note that the closing
parenthesis is missing in the statement funl('Harsh'. The code is followed by
the message that appears when executing the code.
Code
def fun1(a):
print('\nArgument\t:',a)
print('\nType\t:',type(a))
fun1(34)
fun1(34.67)
fun1('Harsh'
The second type is more complex. At times the program stops working or
behaves in an undesirable way on execution. This may be due to incorrect user
input, inability to open a file, accessing something which the program does not
have authority to do and so on. These are referred to as exceptions. Exceptions
are “events that modify the follow of the program”[1]. Python invokes these
events when errors occur, or the programmer can explicitly invoke them.
Exceptions are used to handle some situations. So if something undeniable

comes up, the programmer must have a place to go (part of the code) where that
situation can be handled. In order to understand the point, consider the following
example.
Suppose you intend to design a machine learning technique to identity whether a

given EEG shows an epileptic spike. You decide the algorithm to be used, the
language, and the tool, etc. However, you are not able to get the data. What will
you do? Simply abandon the project and go to the exception handling part. That
is, an exception is raised when situations like the above crop up. Let us consider
one of the most common examples of exception handling. If one is crafting a
program to divide two numbers entered by the user, an exception should be
raised if the denominator entered is zero.
One of the most common ways of handling an exception is to craft a block,

where one expects exception to occur. If somewhere in that block an exception is
raised, the control goes to the part which handles the exception. The block where
you expect the exception to come is the try block and the part where it is
handled is the expect block. The chapter discusses some more ways to handle
exceptions. However, the readers from C++ or C# backgrounds will be familiar
with the above technique and should find this easy and intuitive. Though Python
has a mechanism to handle exceptions, the reader is expected to learn to code his
own classes to handle exceptions. Therefore, the reader must revisit the chapter
on classes and objects.
Exception handling in Python can be done using any of the following:
try/expect
try/expect/finally
raise
assert
This chapter concentrates on the first three. The chapter has been organized as
follows. Section 13.2 discusses the importance and mechanisms of exception
handling, Section 13.3 presents some of the built-in exceptions in Python,
Section 13.4 summarizes the process by taking an example, Section 13.5
presents another example of exception handling and the last section concludes
the chapter.
13.2 IMPORTANCE AND MECHANISM

Exception handling mechanisms can help the programmer to notify something.
For example, consider the problem discussed in the previous section. You have
the EEG of the patients and you want to find the epileptic spike in the EEG. If
you are not able to find the spike, you can simply raise an exception. This
technique is better than the conventional method of returning an integer code on
being able to find something (or for that matter, not find something). Likewise,
on detecting some special case or an unusual condition an exception can be
raised.
At the runtime if an error crops up an exception is raised. This exception can be

handled by the corresponding expect or can be simply ignored. Moreover, if
there is no provision to handle the exception in the code then Python’s default
error handling mechanism comes into play. As stated earlier, on encountering an
error condition the execution is restored after the try statement.
Python also has the try/finally statements to handle the termination condition.
Those of you from a Java background will be familiar with finally. It is for
handling the termination condition, whether or not an exception has occurred.
For example in designing software the concluding screen must appear, whether
or not an exception has occurred or for that matter the memory of objects must
be reclaimed at the end. For such type of situations finally is immensely helpful.
13.2.1 An Example of Try/Catch
A list contains an ordered set of students. The first location contains the name of
the students who got the highest marks, the second student’s name is at the
second position, and so on.
>>>L = ['Harsh', 'Naved', 'Snigdha', 'Gaurav']
In order to access an element at a given location, the user is asked to enter the
index
>>>Index=input('Enter the index')
Now, the element at that position is accessed using the following statement
>>>print(L[int(index)])
So, if the user enters 1, ‘Naved’ would be the output, if he enters 2, “Snigdha”
would be the output. However, the following message appears if he enters
anything above 3.
The error can be handled using try/catch as shown.
Code
Note that the try block contains the part of the code where the exception may
come up. If a runtime error is there, the except part handles it. Also note that the
except may have the name of the predefined exception. The statement after the
except always executes, whether or not an exception has been raised. The reader
is expected to take note of the fact that the control does not go back to the point
where the expectation really occurred. It can only handle the exception in the
requisite block, after which the normal execution continues. The syntax of the
exception handling mechanism is as follows.
Syntax
try:
##code where exception is expected
expect <Exception>:
##code to handle the exception
## rest of the program
13.2.2 Manually Raising Exceptions

The discussion so far has been concentrated on the situations wherein exceptions
were raised and caught by Python itself. In Python, one can also manually raise
the exceptions. The keyword 'raise' is used to explicitly trigger an exception.
The keyword is followed by the <exception name> (same as that which is
caught). The mechanism of handling such exceptions is the same as described
above. That is, the corresponding expect would handle the thrown exception.
The syntax is as follows.
Syntax
try:
raise <something>
expect <something>:
##code which handles the exception
##rest of the code
If such exceptions are not caught, they are handled in the same fashion as in the
above section. The examples in Section 13.4 present codes where the exceptions
have been raised and caught.
13.3 BUILT-IN EXCEPTIONS IN PYTHON

If the programmer is able to raise specific exceptions, the program will be more
effective. To be able to do so one must know the predefined exceptions in
Python and then use these at appropriate places. This section presents some of
the most common exceptions in Python. The following sections present the use
of these exceptions:
AssertionError
When an assert statement fails, the AssertionError is raised.
AttributeError
When an assignment fails, the AttributeError is raised
EOFError
When the last word of the file is reached and the program attempts to read
any further, the EOFError is raised.
FloatingPointError
This exception is raised when floating point operations fail.
ImportError
If the import statement written in the code cannot load the said module, this
exception is raised. This is same as the ModuleNotFoundError in the later
versions of Python.
IndexError
When the sequence is out of range, this exception is raised.
KeyError
If in a dictionary the key is not found, then this exception is raised.
OverflowError
Note that each data type can hold some value and there is always a
maximum limit to what it can hold. When this limit is reached, the
OverflowError is raised.
RecursionError
While executing a code that uses recursion, at times maximum iteration
depth is reached. At this point in time, the recursionError is raised.
RuntimeError
If an error occurs and it does not fall in any of the said categories, then this
exception is raised.
StopIteration
If one is using the _next_() and there are no more objects that can be
processed, then this exception is raised.
SyntaxError
When the syntax of the code is incorrect, this exception is raised. For
example, not writing the import statements or any such thing.
IntendationError
When incorrect use of indentation is done, then this exception comes up.
TabError
An inconsistent use of spaces or tabs leads to this type of error.
SystemError
If some internal error is found, then this exception is raised. The exception
displays the problem that was encountered due to which the exception is
raised.
NotImplementedError
If an object is not supported or the part that provides support has not been
implemented, then the NotImplementedError is raised.
TypeError
If an argument is passed and is not expected, the TypeError is raised. For
example, in a program that divides two numbers entered by the user, a
character is passed, then TypeError is raised.
ValueError
When an incorrect value is passed in a function (or an attempt is made to
enter it in a variable), the ValueError is raised. For example if a value which
is outside the bounds of an integer is passed then this exception is raised.
UnboundLocalError
This exception is raised when a reference is made to a variable which does
not have any value in that scope.
UnicodeError
This is raised when errors related to Unicode encoding or decoding come
up.
ZeroDivisionError
The division and the modulo operation has two arguments. If the second
argument is zero, this exception is raised.
13.4 THE PROCESS

This section revisits the division of two numbers and summarizes how to apply
the concepts studied so far. Consider a function that takes two numbers as input
and divides those two numbers. If the function is called and two integers are
passed as arguments (say, 3 and 2), an expected output is produced if the second
number is not zero. However if the second number is zero, a runtime error
occurs and an error message (shown as follows) is produced. That is, Python
handles exceptions automatically.
The program can be made user friendly by printing a user friendly, easy to
understand message. This can be done by using exception handling.
Code
def divide(a,b):
result =a/b
print('Result is\t:',result)
divide(3,2)
divide(3,0)
>>>
Output
13.4.1 Exception Handling: Try/Except

The above problem can be handled by using the try/except construct to handle
the run time error. The part of the code where the exception is likely to be raised
is put in the try block. If an exception is raised, it will be handled in the except
block. The except block can have the user friendly error message or the code
which would handle the exception. The following code shows the use of the try
block and displays how a run time error can be handled in the except block.
Note the statement which divides the two numbers is in the try block. If the
second number is zero an exception will be raised and the statements in the
except block will be executed.
Code
Output
======== RUN C:/Python/Exception handling/divide.py ========
Result is : 0.6666666666666666
Exception caught
>>>
13.4.2 Raising Exceptions

One can also raise specific exceptions. For example, the following code raises
the ZeroDivisionError if the second number is zero. Note that the
corresponding except block catches this exception. This can be done if the user
is sure which exception to raise in a given situation. Moreover, there is a chance
that the programmer fails to raise the correct exception thus leading to the
invocation of the automatic exception handling mechanism of Python.
Some of the common exceptions and their meanings have already been
presented in Section 13.3. However, there are many more. The list of such
exceptions can be found at the link provided in the references at the end of this
book.
Code
Output
===== RUN C:/Python/Exception handling/divide raise.py =====
Result is : 0.6666666666666666
Exception caught:ZeroDivisionError
>>>
13.5 CRAFTING USER DEFINED EXCEPTIONS

So far we have seen the automatic exception handling capabilities of Python.
That is, even if there is no try/except, Python handles exceptions. The use of
try/except has also been discussed. The use of raise makes the exception
handling more meaningful, as one can raise specific exceptions as per the needs.
However, so far we have not seen how to deal with the situation, which requires
us to raise a user defined exception. This section discusses the crafting and use
of user defined exceptions.
Suppose there is a situation where a specific exception (as per the need of the
program) is to be raised. However, there is no predefined exception to handle
that situation. In such cases a class which would handle the exception that needs
to be created. The said class should be a subclass of the Exception class, so
that it can be used for raising exceptions. When the situation arises the exception
can be raised, as shown in the following illustration. In the illustration that
follows, a class called My Error, which is derived from Exception, is created.
The _init_ of this class may contain the message which will be printed when
the exception is raised. While raising the exception the keyword raise, followed
by the name of the class is written. The reader is expected to observe the output
and understand that first the message written in _init_ is printed, followed by
the message in the except block. Though this is just a dummy example, it gives
an idea as to how to craft classes that handle exceptions.
Code
Output
====== RUN C:\Python\Exception handling\MyException.py
======
Result is : 0.6666666666666666
My Error type error
Exception caught: My Error
>>>
13.6 AN EXAMPLE OF EXCEPTION HANDLING

The following program finds the maximum number from a given list. The idea is
simple. Initially, the first item is taken as “max.” The items of the given list are
then traversed. While traversing, if any element is greater than that stored in
“max” then that number is stored in the variable “max.” At the end the value of
“max” is printed. The program and the corresponding output follow.
Code
Output
==== RUN C:/Python/Exception handling/Example/findMax.py
====
Maximum : 89
>>>
Note that if the contents of L are strings (e.g. L=[“Harsh,” “Nakul,” “Naved,”
“Sahil”]), the strings would be compared as per the rules and the largest
(“Sahil”) would be printed. That is the program works for integers, strings or
floats. However, for the following list an exception would be raised:
L= [2, 'Harsh', 3.67]
Output
The problem can be handled by putting the part of the code where the problem is
likely to come in the try block. Moreover, if all the items of the list are to be
entered by the user then the possibility of a runtime error cropping up will be
higher. In such cases the programmer must make sure that everything, including
the input of items and calling the function, are in the try block. The following
code presents the version of the program where items are entered by the user and
exception handling is implemented. Note that the first run produces an excepted
result, whereas the second run results in a runtime error and hence the exception
handling mechanism is invoked.
Code
Output (First run)

Also note that if a 'finally' is added to the code, the statements in finally are
always executed whether or not an exception occurs. The code that contains both
finally and except is presented as follows. Note that the first output produces
the expected result and also prints the statements given in finally. The second
output results in a runtime error and invokes exception handling mechanisms
and also prints the message in finally. The reader is also expected to appreciate
that there was no need whatsoever of the except as finally is already there.
The code would have run correctly, as in handled the runtime error with a
finally, however both have been included to bring home the point that expect
does its intended job with a finally and finally can be used for cleanup
actions or for de-allocating memory and so on.
Code
Output (first run)

>>>
== RUN C:/Python/Exception handling/Example/findMax
exception.py ==
Enter items (press 0 to end)
1
Enter item (press 0 to end)
4
2
89
3
0
List
['1', '4', '2', '89', '3']
Maximum : 89
This is always executed
>>>
Output (second run)

>>>
== RUN C:/Python/Exception handling/Example/findMax
exception.py ==
Enter items (press 0 to end)
3
1
7
harsh
Run time error
This is always executed
>>>
13.7 CONCLUSION
The chapter presents a remarkable way to deal with exceptions. Though Python
has an inbuilt mechanism to deal with exceptions, the knowledge of exception
handling makes programming more effective, user friendly, and robust. The first
step will be to identify the part of the code where exceptions are likely to come,
and put that part in the try block. The exceptions can also be manually caught
and handled in the except block. The finally block handles the unhandled
exceptions and also executes even if there is no exception. The chapter also
presents some of the most common exceptions that can be caught in Python. The
reader is expected to use the concepts learned in this chapter in his/her programs.
Happy programming!
GLOSSARY
try/except :Syntax
try:
##code where exception is expected
expect <Exception>:
##code to handle the exception
## rest of the program
Manually raising exceptions: Syntax

try:
raise <something>
expect <something>:
##code which handles the exception
##rest of the code
POINTS TO REMEMBER
At the runtime, if an error crops up then an exception is raised.
Exception handling in Python can be done using any of the following:
– try/catch
– try/finally
– raise
– assert
In Python, one can also manually raise the exceptions.
The part of the code where the exception is likely to be raised is put in the
try block. If an exception is raised, it will be handled in the except block.
The class that helps to raise user defined exceptions should be a subclass of
the Exception class.
The statements in finally are always executed, whether or not exceptions
occur.
EXERCISES
1. Exception handling
(a) Handles runtime errors
(b) Provides robustness in a program
(c) Both
2. Exception handling is needed for

(a) Syntax errors
(b) Run time errors
(c) Both
3. Which of the following is not supported in Python?

(a) Nested try
(b) Re-throwing an exception
(c) Both are supported
(d) None of the above is supported
4. Which of the following is raised in the case of division by zero?

(a) Divide
(b) Zero Divide
(c) Both
5. Which of the following is raised in the case when an index outside the
bounds is accessed?
(a) Array index out of bound
(b) Out of bound
(c) Array

(a) For each try there is exactly one catch
(b) Every try must include a raise
(c) A catch can handle any type of exception
(d) A catch can handle the exception for which it is designed, unless it is
catch all in which case it handles all the exceptions
7. How many “expect’s” can a try have?

(a) Single
(b) Two, only in specific conditions
(c) Any number of catch
8. Which type of exception can be raised?

(a) Predefined
(b) User defined
(c) Both
9. What is the base class of a class, of whose exception is to be raised?

(a) Exception
(b) Error
(c) Both
10. Which is the correct syntax of raise?

(a) raise <name of the exception>
(b) raise(<name of the exception>)
(c) raise(new <user defined exception>)
THEORY
1. What is the difference between compile time and runtime error?
2. What is exception handling?
3. Explain the mechanisms of exception handling.
4. Explain how to create a class that derives from the exception class. How is
this class used to raise exceptions?
5. Explain the function of the following exception classes
PROGRAMMING
The roots of a quadratic equation ax2 + bx + c = 0 are given by the formula
. Write a program to ask the user to enter the values of a, b and
c and calculate the roots.
1. Use try/except in the above question to handle the following situations.

(a) Calculating the root of a negative number
(b) Division by zero
(c) Incorrect format
2. Create a class called negative_discriminant, which is a subclass of the

exception class. Now in question 1 raise the negative_discriminant
exception when the value of b2 – 4ac is negative.
The division of two complex numbers is defined as follows. If c1 = a1 + ib1
is the first complex number c2 = a2 + ib2 and is the second complex
number, then the complex number
c = (a1 × a2 – b1b2)/(a22 + b22) + i(a1 × b2 + b1a2)/(a22 + b22)

3. Create a class called Complex and implement exception handling in the
method that carries out division.
4. For the complex class defined in the previous question, use exception
handling to prevent the user from entering a non-real number (as real or
imaginary part).
5. In the complex class, create a function that converts complex numbers to the
polar form.
6. Implement stacks using lists. Incorporate exception handling.
7. Implement queues using lists. Incorporate exception handling.
8. Implement the operations of linked list, and throw an exception when the
number entered by the user is negative. Assume that the data part of the
linked list would contain numbers only.
9. Write a program that takes the ppm of chorine in water from the user and
finds whether the ppm is within permissible limits. If it is not, the program
should raise an exception.
10. Write a program that finds the inverse of a given matrix. The program
should raise an exception when the determinant of the matrix is zero.
14
CHAPTER
INTRODUCTION TO DATA
STRUCTURES
• Understand the importance of data structures
• Classify data structures
• Define stack, queue, tree, and graph
• Define algorithms and appreciate the characteristics of an algorithm
• Understand abstract data types
• Differentiate between iterative and recursive algorithms
• Implement bubble sort, selection sort, merge, and merge sort
14.1 INTRODUCTION
The last two sections introduced procedural and object-oriented programming.
The concepts studied so far constitute the basis of programming. Having learned
the basic concepts, let us move towards becoming a programmer. To become a
programmer the knowledge of data structure is essential. In any project storing
data, its organization and ways to access the data are some of the most important
tasks. The organization of the data comes under what are generally referred to as
data structures. The knowledge of these not only help us to make an efficient
and effective program but also to solve various problems at hand by virtue of
their inherent characteristics. This chapter briefly introduces the concept of data
structures and discusses some of the algorithms of some of the most important
data structures.
We begin with the classification of a data structure. Data structures may be

primary or secondary. Primary data structures are those which are provided by
the language. For example, in C, “int,” “float,” and “char” are primary data
structures. The primary data structures or data types of Python have been
discussed in Chapter 2 of this book.
The secondary data structures are formed from primary data structures. Stacks,
queues, trees, and graphs are some examples of secondary data structures. The
secondary data structures can further be classified as linear and nonlinear. Stacks
and queues are linear data structures whereas trees and graphs are nonlinear data
structures. The classification has been depicted in Figure 14.1.
FIGURE 14.1 Classification of data structures
We begin our discussion with one of the simplest data structures, an array. An
array consists of homogeneous elements at consecutive memory locations.
Arrays are used to store elements which are similar, and accessing many such
elements will take less time if they are stored in consecutive memory locations.
Arrays are an integral part of languages like C, C++, Java, and C. In Python,
however, arrays can be accessed using library. The formation, usage and
applications of multi-dimensional arrays have been discussed in the following
chapter - Chapter 18. Section 14.4 of this chapter discusses the various
algorithms related to arrays in detail.
Definition
Array: An array consists of homogeneous elements at consecutive memory

locations
A stack is a linear data structure, which follows the principle of Last In First Out
(LIFO). In order to understand the concept of stacks, consider opening a
document in Microsoft Word. When one opens a document using the “Open”
dialog and finds the document using the “Browse” window, he cannot close the
“Open” dialog until he closes the “Browse” window. Also the application can be
closed only if one closes the “Open” dialog. The “Browse” window, which was
opened at the end, has to be closed first. This is just like a pile of books, from
which only the book which was kept the end can be taken out. Stacks are used in
sub procedure calls and backtracking. The stacks are characterized by an index
called TOP, which is initially -1. The value of TOP increases as and when the
elements are added in the stack. The algorithms of stack and its applications are
discussed in Chapter 15.
Definition
Stack: A stack is a linear data structure that follows the principle of Last In First
Out (LIFO).
Queue is a data structure that follows the principle of First In First Out (FIFO).
Customer services use the idea of Queue. The person who enters first is served
earlier than the person who entered later. When we give many print commands
to a printer, the commands are stored in a queue and the corresponding
documents are printed in the order in which the command was given. Queues are
used in scheduling algorithms by the operating systems, in spooling (e.g. for
printer) and inapt many other places. The queues (static) are characterized by
two indices - REAR and FRONT. The initial value of both the indices is -1. The
value of REAR increases as elements are inserted in the queue and that of
FRONT increases when elements are removed from the queue. The algorithms
of insertion and deletion in a queue are presented in Chapter 15. The chapter also
presents some of the important applications of queue.
Definition
Queue: A queue is a linear data structure that follows the principle of First In
First Out (FIFO).
A graph may be defined as a set containing two finite sets: the set of vertices (V)
and the set of edges (E). They are represented using two-dimensional arrays or
linked lists. A graph can be weighted, in which case the corresponding matrix
would have weights at the requisite positions. Graphs are traversed using
methods like breadth first search or depth first search. Interestingly, these
traversals use linear data structures like stacks and queues. Graphs are
extensively used in many computational problems. In fact, there is a dedicated
branch of algorithms for solving problems using graphs. The concept of
algorithms and applications of graphs are discussed in Chapter 17 of this book.
Definition
Graph: A graph may be defined as G = (V, E), where V is a finite, non-empty
set of vertices and E is a finite, non-empty set of edges. Each element of E is (x,
y), where x and y belong to the set of vertices.
A tree is a graph which does not have any cycle or isolated vertex (or edges). A
tree is a graph and therefore it has vertices and edges. The absence of cycles and
isolated vertices makes it usable for many problems like searching, finding
complexity, and so on. A tree is generally used for representing a hierarchical
relationship. A special tree called a binary tree has only a maximum of two
children at each level. Trees have been discussed in Chapter 17 of this book.
Definition
Tree: A tree is a graph that does not have any cycle or isolated vertex or edge.
In the following chapters, the concepts related to trees and graphs have been
discussed. Trees are graphs and fall under the category of non-linear data
structures. As stated earlier these are extensively used in searching, sorting,
finding minimum spanning trees, and solving some of the most important
problems in the field of computer science.
As well as the above there are other data structures like Plex, which are beyond
the scope of this book. The concept of file has already been discussed in chapter
9 of this book. The organization of a file is a fascinating topic used in many
fields, including data base management system.
The chapter has been organized as follows. This section has introduced the
definitions of a data structures and their types. The definitions of linear data
structures like stacks and queues and nonlinear data structures like graphs and
trees have also been presented in this section. The second section discusses
Abstract Data Types (ADT) and the third section discusses the definition of
algorithms. Section 14.4 discusses arrays and Section 14.5 discusses iterative
and non-iterative algorithms. The topic has been discussed by taking examples
of three popular sorting techniques: selection sort, bubble sort, and merge sort.
The last section concludes the chapter.
14.2 ABSTRACT DATA TYPE

Each language has predefined data types like int, float, etc. The operations that
can be applied to these data types are also defined for each type. As stated
earlier, user defined data types are needed so languages support such data types.
Now, the operations that can be applied to these data types must be at least
declared, if not defined for these types. The higher level abstraction makes the
purpose and use of these data types clear.
Abstract Data Types contain elements along with the set of operations to
manipulate the elements. The operations tell us what is to be done, not how it is
to be done. That is, ADTs are not about implementation, as long as it is correct,
but about the tasks. The higher level abstraction of the task is important. As a
matter of fact, it is the designer’s decision as to what operations will constitute
the ADT. There is no hard and fast rule on the number of operations and the type
of operations. An example of ADT is a stack. A stack, as explained earlier, is a
linear data structure that follows the principle of Last In First Out (LIFO). A
stack can be described with the help of the following operations:
push(item): Inserts 'item' at the top of the stack

pop(): Takes an element out of the stack
isfull(): Returns true if the stack is full
isempty(): Returns true if the stack is empty
overflow(): Returns true if an overflow exception is raised
underflow(): Returns true if an underflow exception is raised
Note that in the above description, the various functions are clear about what is
to be done not how it is to be done. Other examples of ADTs include queues,
lists, trees, graphs, etc. The reader may also note that languages like C++, Java,
C#, Python, etc., support ADTs via classes.
The implementation of an ADT requires data structures. The choice of data

structure, therefore, becomes one of the most contentious issues. An appropriate
data structure must be chosen as per the problem at hand, keeping in view the
efficiency part too.
For example, the stack data structure can be implemented using data structures
like:
Array
Linked list
Having studied the definitions of data structures, let us shift our focus to the way
a problem is solved. The next section introduces the notion of algorithms and the
following section presents arrays.
14.3 ALGORITHMS
In order to accomplish any task, one needs to plan the chain of action. For
example you are required to complete a task consisting of 4 subtasks, where each
subtask can be performed only on the completion of the previous subtask. You
would probably perform the first subtask, followed by the second, followed by
the third and finally the fourth. An example of such a process follows. In the
process, it is assumed that the person has ingredients required for the Indian
recipe (of kheer) that follows.
Ingredients: 1 L milk, ¼ cup rice, 6 tablespoon sugar, some dry fruits (almonds,
raisins, cashews).
The steps to make Indian “kheer” have been depicted in Figure 14.2.
FIGURE 14.2 Making Kheer
A set of steps required to accomplish a particular task is called an algorithm.

Though there can be many ways of achieving a particular task, the algorithm
designer must look for the most efficient way of doing so; both in terms of space
and time. However, this efficiency must not come at the expense of correctness.
Moreover, each statement of an algorithm must be unambiguous. The
characteristics of a good algorithm can be summarized as follows:
Correctness: An algorithm must produce correct results in all cases or it must

explicitly state for which cases it will not work.
Unambiguous: Each statement of an algorithm must be deterministic.
Efficiency: The time required by the algorithm must be as low as possible, and
the space required must also be also be as low as possible.
Finiteness: The number of steps in a given algorithm must be finite.
The time or space complexity of an algorithm can be stated in terms of

asymptotic functions. These functions give an idea of an algorithm’s
comparative performance. As a matter of fact we are not interested in the exact
equation which relates the time taken by the size of the problem, but only a
function that tells us how the algorithm would behave with larger values of n.
Big Oh: The O (big Oh) notation depicts the upper bound of an algorithm.
Formally, for a function f (t), O(f (t) is defined as follows:
g (t) = O (f (t)), if g (t) ≤ cf (t) for some n ≥ n0
Omega: The Ω(Omega) notation depicts the lower bound of an algorithm.

Formally, for a function f (t), Ω (f (t) is defined as follows:
g (t) = Ω(f (t)), if g (t) ≥ cf (t) for some n ≥ n0
Theta: The θ(big Oh) notation depicts the tight bound of an algorithm. Formally,
for a function f (t), θ (f (t) is defined as follows:
g (t) = θ(f (t)), c1g(t) ≤ f (t) ≤ c2 g(t) for some n ≥ n0
14.4 ARRAYS
An array is a linear data structure that has the same type of elements. The
elements of an array are stored in consecutive memory locations. Those of you
from a C background will have already studied arrays. This section implements
insertion and deletion from an array. An insertion can be done at the beginning,
at the end and somewhere in-between. Likewise, a deletion can be done from the
beginning, from the end and from anywhere in-between. Inserting an element at
the beginning requires all elements to be shifted one position to the right, thus
making way for the “item.” After this, the element “item” is inserted at the first
position and the length of the array in incremented by unity. That is,
##Shift each element of the array one position to the right
arr[0] = item
length = length +1
The first step requires O(n) time and the second and the third would require O(1)
each. Therefore the complexity of inserting an element at the first position is
O(n). Inserting an element at the end requires increasing the length of the given
array by one. The task, thus becomes O(1), as we are keeping track of the
number of elements in the variable “length.”
arr[length] = item
length = length +1
Inserting an element after a certain position requires shifting all elements after
that position by one, followed by putting the element at that position and
increasing the length of the array by one.
#shift all elements after 'position' by one.
arr[pos] = item
length = length +1
The average case complexity of the above task would be O(n). Deleting an
element from the beginning requires shifting all the elements, starting from the
second element, to the right (by one) and then reducing the length of the array by
one. That is,
#shift all the elements to the right (starting from the
second element)
length = length -1
The complexity of shifting all the elements to one position to the right would be
O(n) and hence the complexity of the procedure would be O(n). Deleting an
element after a particular position requires shifting all the elements, starting from
(position +1), to the right (by one) and then reducing the length of the array by
one. That is,
#shift all the elements to the right (starting from position
+1)
length = length -1
The complexity of shifting all the elements to one position to the right would be
O(n) and hence the complexity of the procedure would be O(n). Deleting an
element from the end requires reducing the length of the array by one. This takes
O(1), time. That is,
length = length -1
The task can be done by using functions in C but it requires the use of pointers.
The C program to implement insertion and deletion in an array is as follows.
Inserting an element in a given array, deleting an element from a given array (In C)
Note that in C the modular approach used to accomplish the above task is
complex and intricate. It requires the use of pointers and passing of both the
address of the array and its length in each function. The insertion and deletion of
an element in an array is simple in Python. It requires the use of the array class.
The functions to accomplish different tasks have been presented in Table 14.1.
Table 14.1 Functions for array
Name of the function Task

append Adding an element at the end
insert Adding the element at the specified position. The
function has two arguments: first is the element
and the second is the position
count This counts the number of times the argument is
repeated
pop Takes out the top element from the array
remove Removes the element from a given position
reverse Reverses the order of elements in the array
tostring Converts the given array into a string
from array import array

arr = array('i')
arr.append(3)
arr
Out[4]: array('i', [3])
arr.append(5)
arr
Out[6]: array('i', [3, 5])
arr.insert(1,23)
arr
Out[8]: array('i', [3, 23, 5])
arr.insert(0,32)
arr
Out[11]: array('i', [32, 3, 23, 5])
arr.count(3)
Out[12]: 1
arr.pop(2)
Out[13]: 23
arr
Out[14]: array('i', [32, 3, 5])
arr.remove(3)
arr
Out[16]: array('i', [32, 5])
arr.reverse()
arr
Out[18]: array('i', [5, 32])
arr.tostring()
Out[19]: b'\x05\x00\x00\x00 \x00\x00\x00'
arr.write(file)
------------------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-20-e7f729e1f6ad> in <module>()
----> 1 arr.write(file)
AttributeError: 'array.array' object has no attribute
'write'.
14.5 ITERATIVE AND RECURSIVE ALGORITHMS

In order to understand the difference between an iterative and a recursive
algorithm, we will consider the examples of three algorithms for sorting: bubble,
selection, and merge sort. The first two are examples of iterative algorithms and
the third is an example of a recursive algorithm.
14.5.1 Iterative Algorithms

Iterative procedure is one in which each statement happens one after another.
Linear search, bubble sort, and selection, etc., are examples of iterative
algorithms.
Bubble Sort
In bubble sort the first element is compared with the second, the second with the
third and so on; if the element to be compared is smaller than the first, the
elements are swapped. After the first iteration, the maximum element is placed at
the last position. The process is repeated for the second element and so on. The
process has been presented in the following algorithm:
Analysis
The number of times each statement is executed has been presented in Table
14.2. The first statement is executed (n+1) times and the statements inside the
loop are executed n times. Each time the inner loop executes (n-i) times
therefore making the total number as n(n-i-1). The execution of the if block
depends on the data. Note that the total number of executions has n2 as the
highest degree term. Therefore the complexity of the algorithm is O(n2).
Table 14.2 Analysis of bubble sort

FIGURE 14.3 An example of bubble sort
Selection Sort
In selection sort the first element is compared with the rest of the elements and
whichever is smaller is replaced with the first. This results in the minimum
element coming at the first position. The process is repeated for the second
element and so on. The process has been presented in the following algorithm:
Analysis: Note that the algorithm has a nested loop. The outer loop executes n
times and the inner’s execution makes the complexity O(n2).
The process has been exemplified in Figure 14.3. The first position has 4, it is
compared with the second position and since the second position has 1, the
numbers are swapped. Now, 1 is compared with 6, 2, 8, and 3, one by one. Since
none of these is smaller than 1, the numbers remain at their respective positions.
FIGURE 14.4 An example of selection sort
14.5.2 Recursive Algorithms

Recursive algorithms are those in which the procedure is called in the same
procedure. Recursion requires a base case, the solution of which must be known.
If the base case is not given the algorithms will not terminate, thus resulting in
stack overflow when the program executes. Examples of recursive algorithms
are binary search, merge sort, quick sort, etc.
Merge
Given two sorted lists a1 and a2, having n1 and n2 elements, the merged list is a
sorted list having at most (n1 + n2) elements. The procedure for merge is as
follows.
The two lists are merged as follows. The pointer i is kept at the first index of the
first list and j at the first index of the second list. The resultant array, c, is
initialized to [ ] and pointer k points to the first position of c. If the element at a
is less than that at b, the ith element of a is copied to c, and the pointers i and k
are incremented. If the element at a is greater than that at b, the jth element of b
is copied to c, and the pointers j and k are incremented. If the elements at a and
b are equal, the ith element of a is copied to c, and the pointers i, j and k are
incremented. The process is repeated till i becomes n1 or j becomes n2, after
which the elements of the first array (if i<n1) are copied to c. Otherwise (if
j<n2) the elements of the second array are copied to c.
Merge Sort
Merge sort works as follows. Low points to the first index of the array and high
points to the last index of the array to be sorted. If the value of low is equal to
that of high, there is only one element in the array and hence the array is
returned (an array having a single element is deemed to be sorted). Otherwise the
array is split into two halves, merge sort is recursively applied to the two parts
and the result is merged using the procedure explained above.
Analysis
The complexity of merge is O(n) (note that there are loops which run one after
another). The merge sort divides the array into two parts until a single element
array is formed. Therefore, if the number of elements is initially n, after the first
iteration the two parts will have n/2 elements each. In the next iteration, the
number of elements in the two parts will be n/4 and so on. The process
terminates if the number of elements becomes 1. That is . That is i = log2
n. The complexity of merge sort becomes O(n logn).
14.6 CONCLUSION
One of the most important things while designing a program is to do a task
efficiently. The efficiency must be both in terms of space and time. Managing
time complexity requires developing better algorithms whereas the space
complexity can be managed with various factors. This chapter introduces the
concept of data structures for efficient storage, management and access of data.
The classification of algorithms has been discussed and definitions of various
data structures have been presented. The chapter also introduces the concept of
complexity. The examples of iterative and recursive algorithms have been given
to explain the concept. The reader is expected to implement the concepts
presented here. The following chapters take the discussion forward and explain
various applications of stacks, queues, and implementation of linked lists. To
conclude “Bad programmers worry about the code, good programmers worry
about the data structures.”
GLOSSARY
Array: An array consists of homogeneous elements at consecutive memory
locations.
Stack: A stack is a linear data structure that follows the principle of Last In First
Out (LIFO).
Queue: A queue is a linear data structure that follows the principle of First In
First Out (FIFO).
Graph: A graph may be defined as G= (V, E), where V is a finite, non-empty set
of vertices and E is a finite, non-empty set of edges. Each element of E is (x, y),
where x and y belong to the set of vertices.
Tree: A tree is a graph that does not have any cycle or isolated vertex or edge.
Big Oh: The O (big Oh) notation depicts the upper bound of an algorithm.
Formally, for a function f (t), O(f (t) is defined as follows.
g(t) = O(f (t)), if g(t) ≤ c f (t) f or some n ≥ n0
Omega: The (Omega) notation depicts the lower bound of an algorithm.

Formally, for a function f (t), (f (t) is defined as follows.
g(t) = Ω(f (t)), if g(t) ≥ c f (t) f or some n ≥ n0
Theta: The θ(big Oh) notation depicts the tight bound of an algorithm. Formally,
for a function f (t), θ(f (t) is defined as follows.
g(t) = θ(f (t)), if c1g(t) ≤ f (t) ≤ c2g(t) f or some n ≥ n0

POINTS TO REMEMBER
Algorithms should have the following features.
(a) Correctness
(b) Unambiguous
(c) Efficiency
(d) Finiteness
An array is a linear data structure that has the same type of elements.
Iterative procedure is one in which each statement runs after another.
Linear search, bubble sort, selection sort, etc., are examples of iterative
algorithms.
The complexity of bubble sort is O(n2).
The complexity of selection sort is O(n2).
The complexity of merge sort is O(n log n).
EXERCISES
1. Which of the following is correct?

(a) A data structure is used to organize elements.
(b) Data structures may be used to access elements efficiently.
(c) The knowledge of data structures helps the programmer to make
efficient programs.
(d) All of the above.
2. Which of the following is an example of a basic data structure?

(a) int
(b) float
(c) char
3. Which of the following is a linear data structure?

(a) Queue
(b) Stack
(c) Tree
4. Which of the following is an example of a nonlinear data structure?

(a) Tree
(b) Graph
(c) Both
5. A linear data structure which follows the principle of First In First Out is
(a) Queue
(b) Stack
(c) File
6. A linear data structure which follows the principle of Last In First Out is
(a) Queue
(b) Stack
(c) File
7. Which queue is more efficient with respect to the available space

utilization?
(a) Linear queue
(b) Circular queue
(c) Both are equally efficient
(d) No basis for comparison
8. A queue can be used in which of the following applications?

(a) Round robin scheduling
(b) Spooling
(c) Customer services
9. Which search uses queue?

(a) Depth first search
(b) Breadth first search
(c) Both
10. Which search uses stack?

(a) Depth first search
(b) Breadth first search
(c) Both
11. A tree
(a) May have an isolated edge.
(b) May have a loop.
(c) May have an isolated vertex.
(d) None of the above.
12. A graph
(a) May have a loop
(b) May have an isolated edge
(c) May have an isolated vertex
THEORY
1. What is a data structure? Explain the importance of data structures.
2. Classify data structures on the basis of basic and non-basic. Give examples
of each type.
3. Classify data structures on the basis of linear and non-linear. Give examples
of each type.
4. What is a queue? Write an algorithm for the static implementation of a

queue.
5. What is a stack? Write an algorithm for the static implementation of a stack.
6. State a few applications of stack. Explain how a stack can be used in blind
search.
7. Define a tree. What is the difference between a graph and a tree?
8. What is a graph? State a few applications of a graph.
9. Differentiate between iterative and recursive algorithms. Give an example of

each.
10. What is an abstract data type? Explain with the help of an example.
11. What is an array? Write algorithms for the following.

(a) Inserting an element at the beginning
(b) Inserting an element at the end
(c) Inserting an element after a given element
(d) Deleting an element from the beginning
(e) Deleting an element from the end
(f) Deleting a given element from an array
12. Explain bubble sort. What is the complexity of bubble sort?

13. Suggest a change in bubble sort, so that the complexity can be improved.
14. Explain selection sort. What is the complexity of selection sort?
15. Suggest a change in selection sort, so that the complexity can be improved
to O(n log n).
16. Write an algorithm to merge to sorted arrays. State its complexity.
17. Write an algorithm to implement merge sort. What is its complexity?
18. Which is better - bubble sort or selection sort?
19. Which is better - selection sort or merge sort?
20. Which is better in terms of space complexity - merge sort or selection sort?
1. Write a program to implement a stack.
2. Write a program to implement a queue.
3. Write a program to carry out the following operations in an array

(i) Inserting an element at the beginning
(ii) Inserting an element at the end
(iii) Inserting an element after a given element
(iv) Deleting an element from the beginning
(v) Deleting an element from the end
(vi) Deleting a given element from an array
4. Write a program to implement bubble sort. What is the complexity of bubble

sort?
5. Write a program to implement selection sort.
6. Write a program to implement to merge to sorted arrays. Write a program to

implement merge sort.
7. Write a program to concatenate two given arrays.
8. Write a program to find the maximum element from a given array.
9. Write a program to find the second maximum element from a given array.
10. Write a program to reverse the order of elements of a given array.

15
CHAPTER
STACKS AND QUEUES

• Understand the importance of stacks and queues
• Using dynamic tables for implementing stacks
• Understand postfix, prefix, and infix expressions
• Convert infix to postfix, infix, to prefix, and postfix to infix
• Understand the applications of stacks and queues
15.1 INTRODUCTION
Stacks and queues were introduced in the previous chapter. This chapter takes
the topic forward and explains various implementations and applications of
stacks and queues. The data structures are important as they find applications in
recursive algorithms, conversion and evaluation of expressions, operating
systems, and in popular CPU scheduling algorithms like round robin, etc. It may
be stated here that Python provides libraries for both stacks and queues.
However, a programmer is expected to know the implementations of the said
data structures and to understand the advantages and problems associated with
various implementations.
This chapter also introduces infix, postfix, and prefix expressions. Conversion
and evaluation of expressions have been discussed in detail in this chapter. This
topic finds applications in compiler design too. The linked list based
implementation of stacks and queues have been deferred to the next chapter, as
linked lists have been formally introduced in the next chapter.
This chapter has been organized as follows. Section 15.2 presents the basic
terminology and array based implementation of stack. Section 15.3 discusses one
of the dynamic implementations of stack and the discussion continues in the next
section. Section 15.5 discusses two important applications of stacks: reversal of
string and infix, postfix and prefix expressions. Section 15.6 presents the basics
of queues and its implementation. The last section concludes the chapter. The
reader is expected to revisit Chapter 14 before proceeding any further.
15.2 STACK
A stack is a linear data structure that follows the principle of Last In First Out
(LIFO) or First In Last Out (FILO). A stack can be implemented using an array
which has a fixed capacity (say n). The TOP denotes the position at which the
element is to be inserted. Initially, the value of TOP is -1. As elements are
inserted onto the stack, the value of TOP is increased by one until its value
becomes (n-1), after which the 'Overflow' exception is raised. The insertion in
a stack is referred to as 'push'. The algorithm for push has been presented as
follows (Algorithm 15.1).
Algorithm 15.1
An element is deleted from the top of the stack if the value of TOP is not -1, in
which case an underflow exception is raised. Otherwise, the element at TOP is
returned and the value of TOP is decremented by 1. This operation is referred to
as 'pop'. The algorithm for pop is as follows (Algorithm 15.2).
Algorithm 15.2
The push and pop operations take O(1) time each. Supposing all the placeholders
of the given stack are full and each element is to be popped, the total time would
be n × O(1) = O(n).
The above implementation is static implementation. The following illustration

implements the above algorithm (Illustration 15.1).
Illustration 15.1: Write a program to implement a stack.
Solution: The theory has already been discussed.

Output
>>>
============ RUN C:\Windows\System32\stack.py ============
TOP = 0
TOP = 1
TOP = 2
TOP = 3
TOP = 4
Overflow...
TOP= 3
21
TOP= 2
1
TOP= 1
4
TOP= 0
2
TOP= -1
3
>>>
15.3 DYNAMIC IMPLEMENTATION OF STACKS

The problem with the above implementation is that as soon as the maximum
limit is reached (of the maximum number of elements that the stack can have),
overflow occurs. One of the ways of handling the problem is to increment the
size of the stack by 1 as and when a new element is inserted after the limit is
reached. However, this is not a very appropriate solution of the problem as each
time a new array is created the values of the previous arrays are copied into the
new array. So, for the first item a new array is created (initial array). The second
insertion increases the size of the stack by 1 and copies the previous item in the
new stack. This implies that there would be one copy operation and 1 insertion
in inserting the second element. In the third insertion, there would be two copy
operations and 1 insertion. Note that in the nth insertion, there would be (n-1)
copy operations and 1 insertion operation. In total there would be O(n2) copy
operations.
Output
====== RUN C:/Python/Data Structure/Stack_dynamic1.py ======
TOP = 0
TOP = 1
TOP = 2
TOP = 3
TOP = 4
TOP = 5
TOP= 4
71
TOP= 3
21
TOP= 2
1
TOP= 1
4
TOP= 0
2
>>>
15.4 DYNAMIC IMPLEMENTATION: ANOTHER WAY

The problem with the above implementation can be handled by doubling the size
of the array after the limit is reached. This solution is better than other solutions
where the size of the array is incremented by 1, as the number of copy operations
using this method is O(n). So, for the first item, a new array is created (initial
array). The second insertion increases the size of the stack by 1 and copies the
previous item into the new stack. This implies that there will be one copy
operation and 1 insertion in inserting the second element. In the third insertion, a
new stack of size 4 is created and there would be two copy operations and 1
insertion. In the fourth insertion, there is no need to create a new stack. The
reader is expected to carry out the mathematical analysis. Note that this gives
you an idea of amortized analysis. Those of you interested may refer to the links
at the end of this book. The following code presents the implementation.
Code
The implementation of stacks using linked lists has been dealt with in the next
chapter. We now move on to some of the applications of a stack.
15.5 APPLICATIONS OF STACKS

Stacks can be used to carry out a variety of tasks like reversing a string,
evaluation of a postfix expression, conversion of infix to postfix, and evaluation
of postfix. Let us begin with reversing a string of characters.
15.5.1 Reversing a String

A string can be reversed using a stack by adopting the procedure that follows.
From a given string one character is taken at a time and put in the stack. When
all the characters are over, we start popping out characters from the stack. For
example if the input string is “harsh,” the process of reversing the string using a
stack has been depicted in Figure 15.1.
FIGURE 15.1 A stack can be used to reverse the order of characters of a string
Illustration 15.1: Ask the user to enter a string and reverse the string using a
stack (use a list as stack).
Solution: The theory has already been discussed. The program is as follows.
Code
str= input('Enter a string\t:')
rev_string=''
a=[]
for i in str:
a.append(i)
i=0
while i<len(str):
x=a.pop()
rev_string=rev_string+ x
i=i+1
print(rev_string)
Output
>>>
========== RUN C:/Python/Data Structure/reverse.py
==========
Enter a string :harsh
hsrah
>>>
Illustration 15.2: Reverse a line entered by the user using stacks. The stack
need not be implemented. You can use a list as a stack.
Solution: The process remains same. However, to split a line into words the
split() function is used.
line=input('Enter a line\t:')
a=[]
rev_line=''
words=line.split()
print(words)
for i in words:
a.append(i)
i=0
while i<len(words):
rev_line+=a.pop()
rev_line+=' '
i+=1
print(rev_line)
Output
>>>
========== RUN C:/Python/Data Structure/revline.py
==========
Enter a line :I am Harsh
['I', 'am', 'Harsh']
Harsh am I
>>>
15.5.2 Infix, Prefix, and Postfix Expressions

Another important application of stack is to convert an infix expression to that in
postfix and prefix. In order to understand this, let us first see what postfix,
prefix, and infix expressions are. When a binary operator is between two
operands, the expression is referred to as infix. If the operator is after the two
operands, the expression is in the postfix form. If the operator is before the
operands, the expression is in the prefix form. For example, the addition of “a”
and “b” can be written as follows, in different forms:
Infix: a + b
Postfix: ab +
Prefix: + ab
The evaluation of a postfix expression can be done by employing the following

procedure.
Evaluation of Postfix
Step 1: Initialize the postfix expression, P by NULL and let the stack be initially
empty.
Step 2: For an incoming symbol, s, in expression E
Repeat the following steps till there is a symbol in the given string
If it is a operand, put in the stack
Else if it is an operator pop two symbols from the stack (say x; and y, in
that order)
Apply the operator as x + y, and put the result in the stack.
In order to understand the above procedure, let us consider an
expression E
E = ab + c × d/
The steps of evaluation of this postfix expression have been depicted in Table
15.1, shown as follows.
Table 15.1 An example of conversion of a postfix expression to that in

prefix
Symbol Stack Processing

a A push(a)
b a, b push(b)
+ . x = pop(), y = pop(), x+y
. (a+b) push((a+b))
c (a+b), c push(c)
X . x = pop(), y = pop(), x+y
. c X (a+b) (push((c X (a+b)))
d (c X (a+b)), d push(d)
/ . x = pop(), y = pop(), x+y
. (c X (a+b))/d) .push((c X (a+b))/d)
We now come to the conversion of an infix expression to that in postfix. The

conversion of an infix expression to that in postfix can be carried out by
employing the following procedure.
Postfix Conversion
Put a closing parenthesis at the end of the given expression E.

Put an opening parenthesis at the top of the stack.
Initialize P to NULL.
Repeat the following steps until there is a symbol remaining in E.
For each symbol, s.
If “s” is an operand, put it in P
If “s” is an opening parenthesis, put in the stack
If “s” is an operator,
Put “s” in stack, if the top of the stack contains an operator having a lower
priority operator (or for that matter, no operator).
Otherwise pop the topmost symbol from the stack, put it in P and push the
incoming operator in the stack.
If “s” is closing parentheses, continue popping symbols from the stack until
an opening parentheses is found (including the opening parentheses).
The conversion of an infix expression to that in postfix is similar. However, the

given string needs to be reversed first. This is followed by the application of the
above procedure, after which the resulting string should be reversed. The
conversion of an infix expression to that in prefix can be carried out by
employing the following procedure.
Prefix Conversion
Reverse the given expression E, call it E’.

Put a closing parenthesis at the end of the given expression E’.
Put an opening parenthesis at the top of the stack.
Initialize P to NULL.
Repeat the following steps till there is a symbol remaining in E.
For each symbol, s.
If “s” is an operand, put it in P
If “s” is an opening parenthesis, put in the stack
If “s” is an operator,
Put “s” in stack, if the top of the stack contains an operator having a lower
priority operator (or for that matter, no operator).
Otherwise pop the topmost symbol from the stack, put it in P and push the
incoming operator in the stack.
If “s” is a closing parenthesis, continue popping symbols from the stack
until an opening parenthesisis found (including the opening parenthesis).
Reverse the output obtained P, call it P’.
15.6 QUEUE
As stated in the first section, a queue is a linear data structure that follows the
principle of First In First Out (FIFO). A queue is characterized by FRONT and
REAR. Initially, the value of FRONT and REAR are both -1. When a new element is
added to a queue, the value of REAR is incremented, if the value of REAR is not
(n-1). When an element is deleted from the queue, the value of FRONT is
incremented by 1 if FRONT is not -1. In the first insertion REAR and FRONT are both
incremented and the value of REAR (and FRONT) become 0. In the case of deletion
where the value of REAR and FRONT are same, the value of REAR and FRONT
become -1. The algorithms for insertion in a queue and deletion from a queue are
as follows:
The algorithm (both places) for the deletion from a queue is as follows. The
algorithm (both places) returns the value which is deleted from the queue. If a
“-1” is returned, it indicates an underflow.
The following illustration depicts the static implementation of a queue.
Illustration 15.3: Implement queue using a list.
Solution: In the program that follows, the queue has been implemented using a
list called “a.” The initial values of FRONT and REAR are both -1.
Output
15.7 CONCLUSION
Many researchers consider data structures more important than the procedure.
Stacks and queues are the most important aspects of these important things. The
implementations of stacks and queues have been discussed in this chapter. The
conversion of a postfix expression to that in infix, infix to postfix, infix, to prefix
have not been implemented and have been left as an exercise for the reader. The
reader is also expected to visit the links provided at the end of this book to find
some more applications of stacks and queues and try to implement them. Lastly,
the linked list based implementation has been discussed in the next chapter.
GLOSSARY
Stack: A linear data structure which follows the principle of Last in First out
(LIFO).
Queue: A linear data structure which follows the principle of First in First out
(FIFO).
IMPORTANT POINTS
Complexity of insertion and deletion in both stack and queue is O(1).
If the value of TOP is -1, stack is empty.
In the static implementation, if the value of TOP is (n-1), the stack is full,
where n is the maximum number of elements a stack can hold.
On inserting an element, in a stack the value of TOP increases by 1.
On removing an element from a stack, the value of TOP decreases by -1.
Evaluation of postfix expression, conversion of infix to postfix, and that to
prefix requires stack.
Applications of queues include customer services, round robin, printer
spooling, etc.
EXERCISES
1. Books are kept one over another such that the book which is kept at the end
would be picked up first. Which data structure resembles this structure?
(a) Stacks
(b) Queues
(c) Graphs
(d) Trees
2. What is the time complexity of a POP operation in the static implementation

of stack?
(a) O(1)
(b) O(n)
(c) O(n2)
3. What is the time complexity of a PUSH operation in the static

implementation of stack?
(a) O(1)
(b) O(n)
(c) O(n2)
4. What is the space complexity of a PUSH operation in the static

implementation of stack?
(a) O(1)
(b) O(n)
(c) O(n2)
5. In the dynamic implementation of stack which data structure can be used?

(a) Graphs
(b) Trees
(c) Linked lists
6. Which one is more flexible - static or dynamic implementation of stack?

(a) Static
(b) Dynamic
(c) Both of the above
(d) Cannot determine
7. What is the prefix form of the expression: ((a-b) × c)/d.

(a) /×-1bcd
(b) –abcx/d
(c) /×-abcd
8. What is the postfix form of the above expression?

(a) ab-c × d/
(b) ab-cd × /
(c) abc+-/

(a) The prefix of an expression is just the reverse of the postfix
(b) Conversion of an infix expression to postfix and that to prefix follows
similar procedures
(c) Both the statements are true
(d) None of the statements is true
10. Evaluation of postfix requires which of the following data structures?

(a) Stacks
(b) Queues
(c) Graphs
(d) Trees
11. Customer services resemble which of the following data structures?

(a) Stacks
(b) Queues
(c) Graphs
(d) Trees
12. Round robin algorithm requires which of the following data structures?
(a) Stacks
(b) Queues
(c) Graphs
(d) Trees
13. What is the time complexity of the addition of an element in the static
implementation of a queue?
(a) O(1)
(b) O(n)
(c) O(n2)
14. The dynamic implementation of queue requires which of the following data
structures?
(a) Graphs
(b) Trees
(c) Linked lists
15. Which of the following is an application of circular queue?

(a) Traffic lights
(b) Memory management
(c) Adding large integers
THEORY
1. Write an algorithm for static implementation of a stack. What are the
problems in the implementation?
2. State any two methods to address the above problems.
3. State any two applications of stack. Implement any one.
4. Write an algorithm to implement a queue from two stacks.
5. State a few applications of a queue. How is a queue useful in scheduling

using a round robin algorithm? Explain.
6. What is a circular queue? Write an algorithm for implementing a circular

queue.
7. What is a doubly ended queue?
8. Write an algorithm for converting an infix expression to a postfix

expression.
9. Write an algorithm for converting an infix expression to a prefix expression.
10. Write an algorithm for converting a postfix expression to an infix

expression.
NUMERICAL
1. Evaluate the following postfix expressions:
(a) ab – c ×
(b) ab – cd / ×
(c) ab + cd / – f +
(d) ab · c –
(e) xSin
2. Convert the following to postfix:

(a) (a + b) – (c / d ) × f
(b) (a / b) × (c + d)
(c) a /(b + c)) – d
(d) (c · d) × ((a + b) / f )
(e) a + ((b / c) × (d · f ))
3. Convert the following into prefix:

(a) a – ((b / d) × f )
(b) a / (b × (c + d))
(c) (b /(a + c)) × (d – f )
(d) c · ((d × (a – b)) / f )
(e) (a + b) / c × (d · f )
PROGRAMMING EXERCISES
1. Write a program for static implementation of a stack.
2. Write a program to implement a queue from two stacks.
3. Implement a round robin algorithm.
4. What is a circular queue? Write a program for implementing a circular

queue.
5. Write a program for converting an infix expression to a postfix expression.
6. Write a program for converting an infix expression to a prefix expression.
7. Write a program for converting a postfix expression to an infix expression.
USEFUL LINK
1. Lecture notes: https://www.cs.cmu.edu/~rjsimmon/15122-s13/09-
queuestack.pdf.
CHAPTER
16
LINKED LISTS
• Understand the need and importance of linked lists
• Insert and delete an item in a given linked list
• Implement stack and queue using a linked list
• Understand the problems associated with linked lists
16.1 INTRODUCTION
The previous chapter introduced two of the most important data structures;
namely stack and queue. Stacks follow the principle of Last In First Out (LIFO)
whereas queues are linear data structures which follow the principle of First In
First Out (FIFO). That is, an element can be added or removed only from a
specific position in these data structures. The data structure introduced in this
chapter is far more flexible in terms of the insertion and removal of elements.
The previous chapter also discussed the static implementation of stacks and
queues. But the discussion on the dynamic implementation using linked lists was
deliberately delayed. The linked list discussed in this chapter will help the user
in the dynamic implementation of stack and queue. Here, it may be stated that
Python provides functions for creation of linked lists and supported operations.
The purpose of this chapter is to make the user familiar with the mechanism of
the operations.
Linked list is a data structure whose basic unit is node. Each node has two parts
namely: DATA and LINK. The DATA part contains the value whereas the LINK
part has the address of the next node (Figure 16.1).
FIGURE 16.1 The basic unit of a linked list is a node. A linked list may have any number of
nodes. The last node has NULL in its LINK part.
Linked lists are used in many problems like the implementation of dynamic
stacks and queues and implementing non-linear data structures like trees and
graphs. It may be stated here that although the use of linked lists makes the
dynamic implementation easy, they come with their own problems. In some
cases linked lists are not used in the implementation of trees. In implementing
non-linear data structures the need of linked lists must be deliberated upon. For
example, if we have a balanced binary tree or a heap then the use of arrays is
much better than a linked list.
A linked list consists of nodes connected together via LINK. As stated earlier,
each node has data and a LINK, where data may be a primary or even a
secondary data structure and the LINK is a pointer to the next node. The first
node of a linked list will henceforth be denoted by HEAD. It may also be noted
that the LINK of the last node is NULL, indicating that there is no node after the
last node. Therefore it becomes easy to identify the first and the last node in a
linked list. The following operations can be carried out in a linked list:
1. Insertion at beginning
2. Insertion at middle
3. Insertion at end
4. Deletion at beginning
5. Deletion at middle
6. Deletion at end
The following section explains each of the above operations in detail. Each
algorithm is followed by a figure, which exemplifies the procedure explained in
the example.
16.2 OPERATIONS
A linked list can be used in various applications. However, to be able to use
them, one must be equipped with the procedures to insert and delete elements at
the beginning, at the end and at the specified position. The following discussion
throws light on the requisite procedures.
Insertion at beginning
In order to insert a node at the beginning, a new node (say TEMP) is created and
the data is inserted in the DATA part of the TEMP. Now, the NEXT of TEMP
would point to the HEAD of the linked list and finally, TEMP becomes the new
HEAD. The insert_beg(VALUE) presents the algorithm to insert a node at the
beginning and Figure 16.2 presents the implementation of this process.
Algorithm
insert_beg (VALUE)
{
//Create a node called TEMP.
TEMP = node()
//Now put the given value (VALUE) in the data part of TEMP.
TEMP->DATA = VALUE
//Set the LINK part of TEMP to FIRST.
TEMP->LINK = FIRST
Rename TEMP to FIRST.
}
The algorithm has been explained in the following figure (Fig. 16.2).
FIGURE 16.2 Inserting a node at the beginning
Insertion after a particular node
In order to insert a node after a particular node, a node pointer called PTR is
created. Initially PTR points to the HEAD. The LINK of PTR becomes the
LINK of the current node until the current node has the value after which we
intend to insert the new node. A new node called TEMP is created and VALUE
is inserted in its DATA part. Let the LINK of PTR be PTR1. Finally, the LINK of
PTR will point to TEMP and the LINK of TEMP will point to the PTR1. The
insert_middle(VAL, VALUE) presents the algorithm to insert a node after a
given node (having data VAL) and Figure 16.3 shows an example of this
process.
Algorithm
The algorithm has been explained in the following figure (Figure 16.3).
FIGURE 16.3 Insertion after a particular position
Insertion at the end
In order to insert a node at the end, a node pointer called PTR is created. Initially
PTR points to the HEAD. The LINK of PTR becomes the LINK of the current
node till the current node’s LINK is NULL. A new node called TEMP is created
and VALUE is inserted in its data part. Finally, the LINK of PTR will point to
TEMP and the LINK of TEMP will be NULL. The insert_end(VALUE) presents
the algorithm to insert a node after a given node and Figure 16.4 shows an
example of this process.
Algorithm
FIGURE 16.4 Insertion at the end
Deletion From the Beginning
In order to delete a node from the beginning, the LINK of HEAD becomes the
new HEAD. Also, if required, the DATA of HEAD may be stored in some
memory location. The del_beg() presents the algorithm to delete the first node
of the linked list and Figure 16.5 shows an example of this process.
Algorithm
Delete_beg()
{
Set backup = HEAD->DATA
Rename HEAD->LINK as HEAD
}
FIGURE 16.5 Deleting a node from the beginning
Deletion of a node after a particular node (having VALUE = VAL)
In order to delete a node which is after a particular node, the following

procedure is employed. A node pointer PTR is created, which initially points to
the HEAD. The LINK of PTR becomes PTR LINK till the PTR DATA
becomes VAL. If the LINK of this PTR point to PTR1 and PTR1’s LINK points
to PTR2, the LINK of PTR will point to PTR2 and, if required, the DATA of
PTR1 can be saved. The del_middle() presents the algorithm to accomplish the
above task and Figure 16.6 shows an example of this process.
Algorithm
FIGURE 16.6 Deleting a node from a particular position
Deletion of a node from the end
In order to delete a node from the end, the following procedure is employed. A
node pointer PTR is created, which initially points to the HEAD. The LINK of
PTR becomes PTR->LINK till the PTR->LINK->LINK becomes NULL.
Finally, the LINK of PTR will point to NULL. If required the DATA of PTR-
>LINK can be saved. The del_end() presents the algorithm to accomplish the
above task and Figure 16.7 shows an example of this process.
Algorithm
FIGURE 16.7 Deleting a node from the end
Illustration 16.1: Write a program to implement various operations of a singly

linked list.
Code: The algorithm of each operation has already been discussed. The
following code presents the Python implementation of the above code. Note that
the code has 6 functions (in the linked list class), each implementing the
corresponding algorithm.
Output
List
List
2
List
5 2
List
7 5 2
List
7 5 8 2
List
7 5 8 2 9
Length 5
List
5 8 2 9
List
5 8 9
List
5 8
List
Cannot delete
>>>
The above linked list is referred to as a singly linked list. There are other variants
as well, such as a doubly linked list. In a doubly linked list, each node has two
pointers PREVIOUS and NEXT along with the data part. The PREVIOUS
connects the node to the previous node and the NEXT connects the node to the
next node. Also the NEXT of the last node is NULL, indicating that this node is
not connected to any other node. The first node of a doubly linked list is also
given a special name.
For example, if A and B are connected as shown in the figure then the “NEXT”
of A is address of B and the “PREVIOUS” of B is the address of A (Figure
16.8).
FIGURE 16.8 Node of a doubly linked list are connected via two pointers: NEXT and PREV
In a doubly linked list the following operations can be carried out:
1. Insertion at beginning
2. Insertion at middle
3. Insertion at end
4. Deletion at beginning
5. Deletion at middle
6. Deletion at end
There is another variant of a linked list, which is a circular linked list. In a

circular linked list last node is connected to the first node. That is, the “NEXT”
of the last node contains the address of the first node.
16.3 IMPLEMENTING STACK USING A LINKED LIST

A stack, as explained earlier, supports only two operations: insertion at the end
and deletion from the end. The implementation of a stack using a linked list,
therefore, requires only these two operations to be implemented. A linked list
having only two operations insert_end (VAL) and del_end (Section 16.2)
would be a stack. The first operation is equivalent to push and the second is
equivalent to pop(). Illustration 16.2 presents the implementation of a stack
using linked list.
Illustration 16.2: Write a program to implement stack using linked list.

Code
Output
>>>
========== RUN C:/Python/Data Structures/Stack.py ==========
Stack
Stack
2
Stack
2 5
Stack
2 5 3
5 popped
Stack
2 5
2 popped
Stack
2
>>>
16.4 QUEUE USING A LINKED LIST

A queue, as explained earlier, supports only two operations: insertion at the end
and deletion from the beginning. The implementation of a queue using a linked
list, therefore, requires only these two operations to be implemented. A linked
list h aving only two operations Insrt_end(VAL) and del_beg, would be a queue.
The first operation is the same as en_queue and the second is the same as the
de_queue(). Illustration 16.3 presents the implementation of a queue using
linked list.
Illustration 16.3: Write a program to implement queue using linked list.
Code
Output
>>>
========== RUN C:/Python/Data Structures/Queue.py ==========
Queue
Queue
2
Queue
2 5
Queue
2 5 7
Queue
5 7
Queue
7
>>>
16.5 CONCLUSION
A linked list contains connected nodes. Each node has two parts: the DATA part
and the LINK part. The DATA part may contain a basic or a complex data
structure. The LINK part contains the address of the next node. The last node,
therefore, has NULL in its LINK part. One can have linked lists of integers,
float, char, or even strings. The reader is expected to explore the linked lists of
user defined data structures and use them in solving problems. Problems related
to polynomial addition and subtraction have been discussed in the references
given at the end of the book. The reader is expected to go through the theory and
implement the operations using a linked list. Moreover, the algorithms of a
doubly and a circular linked list may be developed in a way similar to that of a
singly linked list given in the chapter. Having studied linked lists and arrays the
reader is also encouraged to find the advantages of using linked lists and also its
disadvantages (for example extra overhead, pointers, etc.). The most important
attribute of a linked list is its flexibility.
GLOSSARY
Node: A node is the basic unit of linked list. It has two parts: data and link. The
link points to the next node.
POINTS TO REMEMBER
The link of the last node of a linked is NULL
Complexities of operations, discussed in the chapter
Operation Complexity
Insertion at the beginning O(1)
Insertion at end O(n)
Insertion in the middle O(n)
O(1)
Deletion from the beginning
Deletion from the end O(n)
Deletion from the middle O(n)
EXERCISES
1. The last element of a singly linked list points to

(a) The first element
(b) NULL
(c) Any element
2. A linked list can

(a) Grow
(b) Shrink
(c) Both
3. In a problem we need a maximum of 100 items; which data structure would

be best suited if only the last element of the data structure is to be accessed?
(a) Stack
(b) Queue
(c) Linked list
(d) Both stack and linked list
4. In a linked list an element can be inserted at

(a) The first position
(b) The last position
(c) At any position
5. Which of the following makes arrays less advantageous when compared to

linked lists?
(a) Array have a fixed size
(b) The elements of an array are stored at consecutive memory locations
(c) It is difficult to add/delete an element from a given position
6. What is the complexity of indexing in a linked list?

(a) O(n)
(b) O(1)
(c) O(n2)
7. What is the complexity of indexing in a dynamic array?

(a) O(1)
(b) O(n)
(c) O(n2)
8. What is the complexity of inserting an element at the end in an array?

(a) O(1)
(b) O(n)
(c) O(n2)
9. What is the complexity of inserting an element at the end in a linked list?

(a) O(1)
(b) O(n)
(c) O(n2)
10. When wasted space is concerned (pointers etc.), what is space complexity in
a linked list?
(a) O(1)
(b) O(n)
(c) O(n2)
an array?
(a) O(1)
(b) O(n)
(c) O(n2)
a dynamic table?
(a) O(1)
(b) O(n)
(c) O(n2)
13. Which is better in terms of time complexity for the operation that adds an
element in the middle?
(a) Array
(b) Linked List
(c) Both
14. Which is better in terms of time complexity for the operation that adds an
element in the middle?
(a) Array
(b) Linked List
(c) Both
15. In a circular linked list the last element points to

(a) The first element
(b) The middle element
(c) NULL
THEORY
1. What is a linked list? Write an algorithm for the following
(a) Inserting an element at the beginning of a linked list
(b) Inserting an element at the end of a linked list
(c) Inserting an element after a given element of a linked list
(d) Deleting an element from the beginning of a linked list
(e) Deleting an element from the end of a linked list
(f) Deleting an element after a given element from a linked list
2. Derive the time complexity of each of the above.
3. What is a doubly linked list? Write an algorithm for the following

(a) Inserting an element at the beginning of a doubly linked list
(b) Inserting an element at the end of a doubly linked list
(c) Inserting an element after a given element of a doubly linked list
(d) Deleting an element from the beginning of a doubly linked list
(e) Deleting an element from the end of a doubly linked list
(f) Deleting an element after a given element from a doubly linked list
4. Derive the complexity for each of the above algorithms.
5. Write an algorithm to implement a stack using a linked list.
6. Write an algorithm to implement a queue using a linked list.
7. Write an algorithm to invert a linked list.
8. Write an algorithm to implement a queue using two stacks.
9. Write an algorithm to find whether a given string is a palindrome.
10. Write an algorithm to find the maximum element in a linked list.
PROGRAMMING
Implement Q1, Q3, & Q5-Q10 of the above section.
EXPLORE
The chapter gives an introduction to linked lists. The reader is expected to read
and implement the algorithms of doubly linked lists and circular linked lists.
For doubly linked lists, the reader may refer to:

https://www.cs.cmu.edu/~guna/15-123S11/Lectures/Lecture11.pdf
CHAPTER
17
BINARY SEARCH TREES

• Understand the terminology and representation of trees and graphs
• Understand the importance of a binary search tree
• Implement insertion, searching, and traversal in a BST
17.1 INTRODUCTION
Consider the hierarchical structure of your college. The college should have the
head under which the deans of various faculties work. The faculty is made up of
various departments, which have their respective heads. These heads take the
help of the chairs of various committees. This heretical structure can be viewed
as a tree with the head of the institute at the Oth level, the deans at the first level,
and the heads of the departments at the second level and so on.
Let us take another example, that of tic-tac-toe. At the beginning, the first player
may fill any of the 9 cells in a 3 × 3 grid. After this move the second user may
fill the other symbol in any of the remaining squares, keeping in mind the
constraints. The game can thus be represented as a tree. Likewise, a tournament
can be represented using a tournament tree.
In the case of machine learning, the decision trees help us to learn. The heap, a
type of tree, helps us to find the maximum or minimum in O(1) time.
There are numerous applications of trees and graphs. This chapter introduces
trees and graphs and concentrates on a specific type of tree called the binary
search tree, which is important in searching and is the basis of many other
important topics.
The chapter has been organized as follows. The second section presents the
definition, terminology, and representation of trees and graphs. The third section
discusses the binary search trees (BST) and the last section concludes the
chapter.
17.2 DEFINITION AND TERMINOLOGY

So far linear data structures, like stacks and queues, have been discussed. The
data structures, their implementations and applications were presented in the
previous chapter. However, there are many applications where nonlinear data
structures are required. This chapter introduces two nonlinear data structures;
namely graphs and trees. However, the focus of this chapter will be on trees -
mainly a specific kind of tree called a binary search tree. Let us begin with the
definition of a graph and its representation.
17.2.1 Graphs: Definition and Representation

Graph: A graph is a set (V, E), where V is a finite, non-empty set of vertices.
The set E consists of tuples (x, y), where x and y belong to the set V. Figure 17.1
shows a graph G = (V, E), where V is (A, B, C, D) and the set E is {(A, B), (A,
D), (B, C), (B, D), (C, D)}.
FIGURE 17.1 A graph G = (V, E), V = {A, B, C, D} and E= {(A, B), (A, D), (B, C), (B, D), (C, D)}
The edges of a graph may have weights, in which case the graph is referred to as
a weighted graph. A graph may be represented using a matrix or a linked list.
The element at the ith row and the jth column in the matrix will be 1, if an edge
exists between the vertex i and the vertex j, otherwise the element will be zero.
In the case of a weighted graph, these elements may represent the weights of the
corresponding edges. The matrix corresponding to the graph of Figure 17.1 is as
follows:
Note that the element at the first row and the second column is 1, as there is an
edge between the first vertex (A) and the second vertex (B). Likewise there is an
edge between A and D, hence the element at the first row and fourth column is 1.
Graphs may also be represented using a set of linked lists, in which each linked
list will point to the vertices connected to the corresponding vertex. The linked
list representation of the graph of Figure 17.1 is as follows (Figure 17.2). Note
that the list of A contains B and D as A is connected to B and D. Likewise, the
list of B contains A, C, and D.
FIGURE 17.2 The linked list representation of the graph presented in Figure 17.1
Graph is one of the most important data structures. This data structure has
applications in a wide variety of tasks like finding the shortest paths, ranking of
web pages, etc. As a matter of fact, there is a dedicated subject on graph
algorithms.
17.2.2 Trees: Definition, Classification, and Representation

Tree: A tree is a non-linear data structure. It is basically a graph which does not
form any cycle and does not have isolated edges or vertices. Figure 17.3 (a)
shows an example of a graph which is not a tree, as there exists a cycle in the
graph. The Figure 17.3 (b) is a tree as it does not contain any cycle or isolated
vertex or edge. The graph of Figure 17.3 (c) is not a tree, as it contains an
isolated edge.
FIGURE 17.3 The examples of graphs, which are not trees ((a) and (c)) and an example of a tree
(b)
Trees can be classified on the basis of the number of children of a node. If each
node of a tree has a maximum of two children, it is called a binary tree. If each
node of a tree has two children except for the last level at which a node has no
child, it is called a complete binary tree. Figure 17.5 shows a complete binary
tree. The root of a tree is always at level 0, the children of the root at level 1 and
so on. The tree of Figure 17.4 is a binary tree, as each node has 0, 1, or 2
children. Note that the node A is at level 0, the nodes B and C are at level 1, and
the nodes D, E, and F are at level 2.
FIGURE 17.4 Binary tree; each node has 0, 1, or 2 children
FIGURE 17.5 Complete binary tree; each node has 2 children except for the nodes at the last
level, at which each node has 0 children
The number of nodes in a complete binary tree having two levels is 3, that
having three levels is 7 and that having four levels is 15. A complete binary tree
having n levels has 2n nodes. In the tree shown in Figure 17.5, A is the root of
the tree as it is at level 0. The nodes B and C are siblings, as they have the same
parent (A). Also, the nodes D and E are siblings and so are F and G. The nodes
D, E, F, and G are the leaves, as they have no children. The following table
(Table 17.1) presents the terminology of a tree.
Table 17.1 Terminology of a tree
Edge: A line which connects two nodes.

Parent: A node from which the given node has been derived.
Root: A node which does not have a parent is called the root.
Degree of a node: The degree of a node is the number of children of a given
node.
Degree of a tree: The degree of a tree is represented by the maximum
number of children.
Level of a tree: The root of a tree is at level 0, the children of root are at level
1 and so on.
17.2.3 Representation of a Binary Tree

A binary tree can be stored in a computer using an array or a linked list. The
array representation of a binary tree requires the root to be stored at the Oth
index. For each node stored at the nth index, its left child would be stored at the
(2n+1)th index and the right child at (2n+2)th index.
FIGURE 17.6 Calculation of index for the array representation of a binary search tree
For example, in the above tree (Figure 17.6) the root would be stored at the 0th
index.
The left child of the root (B) would be stored at the index, given by the
formula
2 × n + 1 = 2 × 0 + 1 = 1
The right child of the root would be stored at the index, given by the
formula
2 × n + 2 = 2 × 0 + 2 = 2
That is B would be stored at the 1st index and C at the 2nd index
Likewise the left child of C would be stored at the index, given by the
formula
2 × n + 1 = 2 × 2 + 1 = 5
The left child of D would be stored at the 11th index and the right child would be
stored at the 12th index. Finally, the left and the right child of F would be stored
at the 25rd and the 26th index. The array representation of the tree of Figure 17.6
is therefore as follows:
As is evident from the above array, this representation suffers from wastage of
space. That is a lot of space is wasted if the given tree is not a completely
balanced tree. Note that in the case of a completely balanced tree, no space
would be wasted in the array representation.
There is another way in which a binary tree can be stored, which is by using a
doubly linked list. In the representation, a node’s left child’s address is stored at
the previous pointer and its right child’s address is stored at the next pointer. The
following figure (Figure 17.7) shows the linked list representation of the tree in
Figure 17.6.
FIGURE 17.7 Doubly linked list representation of a binary tree
17.2.4 Tree Traversal: In-order, Pre-order, and Post-order

A tree can be traversed in in-order, pre-order, and post-order.
In the in-order traversal, the root of a tree is given as an input to the algorithm.
The algorithm works as follows. The left child of the root is given as the input to
the same algorithm (recursion). This is followed by the processing of the root,
following which the right child of the root is given as an input to the same
algorithm (recursion). The algorithm has been presented as follows. The
implementation has been presented in the next section. Though there is a
corresponding non-recursive procedure to accomplish the task, it has not been
discussed in this chapter. The reader may refer to the links at the end of this book
for a more detailed study.
In the pre-order traversal, the root of a tree is given as an input to the algorithm.
The algorithm works as follows. The data of the root is processed first. This is
followed by the output obtained by giving the left child of the root as an input to
the same algorithm (recursion), following which the right child of the root is
given as an input to the same algorithm (recursion). The algorithm has been
presented as follows. The implementation has been presented in the next section.
Though there is a corresponding non-recursive procedure to accomplish the task,
it has not been discussed in the chapter. The reader should refer to the links at
the end of this book for a more detailed study.
In the post-order traversal, the root of a tree is given as an input to the algorithm.
The algorithm works as follows. The output obtained by giving the left child of
the root to the algorithm itself is followed by the output obtained by giving the
right child of the root as an input to the same algorithm (recursion) after which
the root is processed. The algorithm has been presented as follows. The
implementation has been presented in the next section. Though there is a
corresponding non-recursive procedure to accomplish the task, it has not been
discussed in this chapter.
17.3 BINARY SEARCH TREE
One of the advantages of trees is that they help in efficient searches. One of the
variants of binary tree called binary search tree helps to find an element in O(log
n) time (average case). A binary search tree is a binary tree in which each node
satisfies the following property:
The tree shown in Figure 17.8 is a binary search tree, whereas that shown in
Figure 17.9 is not.
FIGURE 17.8 An example of a binary search tree; note that each node’s left child has value less
than the value in the data part and its right child has data greater than the node’s data
FIGURE 17.9 An example of a tree which is not a binary search tree, note that right child of node
having data 10 is less than 10
17.3.1 Creation and Insertion

The creation of a BST is simple. The first value becomes the data part of the
root. To insert a new value in the tree a new node is created, its correct position
is found and the new node is placed at its correct position. The algorithm for
inserting a new node in a BST is as follows.
In order to understand the procedure, let us consider Illustration 17.1. This
illustration is followed by another which implements the algorithm for the
insertion in a BST.
Illustration 17.1: Insert 47, in the BST of Figure 17.10.
FIGURE 17.10 Binary Search Tree for Illustration 17.1
Solution: The process has been depicted in Figure 17.11 (a) to 17.11 (d). The
reader is expected to follow the steps and map them to those given in the
algorithm.
FIGURE 17.11(a) The search begins at the root. Since the value to be searched is greater than
that at root, the right sub tree is searched
FIGURE 17.11(b) The root of the right sub tree is 40. Since the value to be searched (47) is
greater than that at root, the right sub tree of this node is searched
FIGURE 17.11(c) The root of the right sub tree is 51. Since the value to be searched (47) is less
than that at ptr, the left is searched
FIGURE 17.11(d) 47 is inserted
Illustration 17.2: Write a program to insert the value entered by the user in the
binary search tree.
Program
Output
>>>
============ RUN C:\Python\Data Structure\BST.py
============
Tree :
10
Tree :
5 10 20
Tree :
1 2 5 10 15 17 20
>>>
17.3.2 Traversal
As stated in the previous section, a binary tree can be traversed in three ways:
In-order, pre-order, and post-order traversal. The following illustration
implements in-order, pre-order, and post-order traversal of a binary search tree.
Illustration 17.3: Write a program to implement pre-order, post-order, and in-
order traversal of a binary search tree.
Program
Output
>>>
====== RUN C:\Python\Data Structure\Tree Traversal.py ======
In-order Traversal of the BST :
1 2 5 10 15 17 20
Pre-order Traversal of the BST :
10 5 2 1 20 15 17
Post-order Traversal of the BST :
1 2 5 17 15 20 10
>>>
17.3.3 Maximum and Minimum Elements

The maximum element can be found by finding the rightmost element of the
tree. The pointer is first set to the root and iteratively set to the root of the right
sub tree, until ptr reaches terminal node. Logic can be extended to the
complementary problem of finding the minimum element from the tree.
Illustration 17.4: Write a program to find the maximum and the minimum
element from a binary search tree.
Program
Output
>>>
========== RUN C:\Python\Data Structure\BST max.py
==========
Maximum : 20
Minimum : 1
>>>
17.4 CONCLUSION
The chapter introduces one of the most important data structures called a tree.
Since the set of trees is a subset of graphs, the definition and representation of
graphs have also been included in the chapter. Moreover, the programs given in
the chapter use linked list representations of a tree as the array based
implementation becomes inefficient in terms of space if the tree is not a
completely balanced tree. Trees can be traversed in variety of ways. The in-
order, post-order, and the pre-order traversal have been described in the chapter.
One of the most important trees called binary search trees has been introduced in
the chapter. The algorithm and the corresponding program for searching an
element in a binary search tree and insertion have also been included and
exemplified. The reader is expected to visit the links given at the end of the
chapter and explore the algorithm for deleting a node from a binary search tree.
The appendix of this book discusses some of the important graph algorithms
also. Moreover, this is just a beginning of the topic; explore trees and dive into
the exciting world of problem solving through trees.
GLOSSARY
Graph: A graph is a set (V, E), where V is a finite, non-empty set of vertices.
The set E is a set consisting of tuples (x, y), where x and y belong to the set V.
Tree: A tree is a non-linear data structure. It is basically a graph which does not
form any cycle and which does not have isolated edges or vertices.
Edge: A line which connects two nodes.
Parent: A node from which the given node has been derived.
Root: A node which does not have a parent is called the root.
Degree of a node: The degree of a node is the number of children of a given

node.
Degree of a tree: The degree of a tree is the maximum degree of any node of the
tree.
Level of a tree: The root of a tree is at level 0, the children of root are at level 1
and so on.
Binary Search Tree: It is a binary tree in which each node satisfies the
following property:
POINTS TO REMEMBER
The complexity of insertion in a binary search tree is O (log n), if the tree is
balanced.
A tree can be represented using arrays. In this representation, the root is
placed at the 0th index, the right child of a node at the nth index is at (2n +
2)th index and the left child at the (2n + 1)th index.
The linked list representation of a tree is efficient in terms of space.
From a binary search tree, the complexity of finding the
maximum/minimum element is O(log n), if the tree is balanced.
From a binary search tree, the complexity of finding the
maximum/minimum element is O(log n), if the tree is skewed.
EXERCISES

(a) Every tree is a graph
(b) Every graph is a tree
(c) A tree cannot have a cycle
(d) A tree cannot have an isolated edge
2. If G is the set of all the graphs, T is the set of all the trees and BST is the set
of all the binary search trees, then which of the following is false?
(a) G ⊆ T
(b) T ⊆ G
(c) BST ⊆ T
(d) BST ⊆ G
3. What is the depth of a balanced tree having n nodes?

(a) O(n)
(b) O(log n)
(c) O(1)
4. What is the depth of a skewed tree having n nodes?

(a) O(n)
(b) O(log n)
(c) O(1)
5. In which of the following applications trees can be sued?

(a) Searching
(b) Sorting
(c) Priority queues
6. What is the best case complexity of insertion in a binary search tree?

(a) O(1)
(b) O(log n)
(c) O(n)
7. What is the average case complexity of insertion in a binary search tree?

(a) O(1)
(b) O(log n)
(c) O(n)
8. What is the worst case complexity of insertion in a binary search tree?

(a) O(1)
(b) O(log n)
(c) O(n)
9. A BST is created out of sorted list of n numbers; what is the complexity of

inserting an element in the tree?
(a) O(1)
(b) O(n)
(c) O(log n)
10. A balanced BST is given, what is the complexity of inserting an element?

(a) O(1)
(b) O(n)
(c) O(log n)
NUMERICAL
1. Create a BST out of the following list of numbers
(a) 2, 23, 14, 29, 35, 28, 19, 1, 3, 7, 16, 15
(b) 1, 2, 3, 4, 5, 6, 7
(c) 10, 8, 6, 4, 2, 1
(d) 10, 15, 18, 17, 16, 19, 14, 21
(e) 1, 2, 3, 4, 10, 9, 8, 7
2. Which of the above is balanced?
3. Which of the above (Q1) is skewed?
4. The list (b) (Q1) is a sequence; find the nth element of the sequence and the
complexity to insert the nth element
5. In Q1 (d), what is the average complexity of inserting an element?
6. Write the in-order traversal of (a) to (e) (Q1).
7. Write the post-order traversal of (a) to (e) (Q1).
8. Write the pre-order traversal of (a) to (e) (Q1).
9. In Q1, use the in-order and post-order traversal of (a) to recreate the tree.
10. In Q1, use the in-order and pre-order traversal of (b) to recreate the tree.
FIGURE 17.12 A Binary Search Tree

11. Write the in-order traversal of the tree in Figure 17.12.
12. Write the pre-order traversal of the tree in Figure 17.12.
13. Write the post-order traversal of the tree in Figure 17.12.
14. Write the level order traversal of the tree in Figure 17.12.
15. Insert the following in the tree in Figure 17.12 and show the new tree in
each step.
(a) 2
(b) 5
(c) 32
(d) 31
(e) 29
PROGRAMMING
1. Write a program to create a binary search tree from a list of numbers entered
by the user.
2. Write a program to traverse the above tree in

(a) In-order traversal
(b) Post-order traversal
(c) Pre-order traversal
3. Write a program to find a given element from a binary search tree.
4. Write a program to find the maximum element from a binary search tree.
5. Write a program to find the minimum element from a binary search tree.
6. Write a program to find the second maximum element from a given binary
search tree.
7. Write a program to find the sum of elements of a binary search tree.

8. Write a program to find the depth of a binary search tree.
9. Write a program to find the sibling of an element from a binary tree.
10. Write a program to find the parent of an element from a binary tree.
11. Write a program to find all ancestors of a given element of a binary search
tree.
12. Write a program to find all children of a given node of a binary tree.
13. Write a program to delete an element from a given binary tree.
14. Write a program to find the rightmost child of the left sub tree of a node in a
given binary tree.
15. Write a program to find the leftmost child of the right sub tree of a node in a
given binary tree.
CHAPTER
18
INTRODUCTION TO NUMPY
• Understand the downside of Dynamic Typing
• Understand the importance of NumPy
• Create unidimensional and multi-dimensional arrays
• Understand broadcasting
• Understand the importance and creation of structured arrays
18.1 INTRODUCTION
One of the most important features of Python is dynamic typing. In C, the
programmer declares the type of variable and then starts working on it. Though
some amount of typecasting is possible, in general one cannot change the type of
the variable in the program. For example if the programmer declares a variable,
say num, of integer type, and assigns some value to it, he cannot assign a string
(say) to it.
int num;
num=5;
printf("%d", num);
num="Harsh";//error
However, in the case of Python, one needs not declare the type of variable. For
example, in the following code, 5 is assigned to num and the value of num is
printed. In the succeeding statements the string "Harsh" is assigned to num and
the value stored in num is printed. Note that the code executes just fine.
num = 5
print( num )
num = 'Harsh'
print( num )
This is due to dynamic typing, which makes Python stand apart from its other
counterparts. However, this comes at a cost. Each time the variable is assigned a
value the operations that can be applied to that variable must be elucidated,
which takes time. If in a particular situation the type of a variable would not
change, there is hardly any need for such maneuveration. This is also relevant
for procedures which deal with numbers and scientific and statistical
calculations. In such situations, the codes can be made efficient by casting them
to defined types and manipulating them. This chapter throws light on a package
which comes to our rescue in such situations. The package is NumPy.
The chapter has been organized as follows. The second section introduces
NumPy and discusses the creation of a basic array. The section also throws light
on the data types provided by the package. The third section discusses some of
the standard functions for generating sequences. The fourth section discusses
some important aggregate functions. The concept of broadcasting has been
introduced in the fifth section. The sixth section discusses structured arrays and
the last section concludes the chapter.
18.2 INTRODUCTION TO NUMPY AND CREATION OF A

BASIC ARRAY
NumPy is a Python package, which stands for Numerical Python. NumPy contains
multidimensional array objects and the routines which process these arrays. The
package was created by Travis Oliphant. It has almost all the features of
Numeric, its predecessor, and Numarray. One of the greatest advantages of NumPy
(when used along with SciPy, Scientific Python, and MATPLOTLIB) is its ability to
perform operations similar to those in MATLAB. Note that NumPy is open source as
against MATLAB. The package can be easily installed ON your computer from
htpps://www.continuum.io. You can download Anaconda, which is a free
distribution software for SciPy and related packages.
Using NumPy, multidimensional arrays can be subjected to mathematical and

logical operations, transforms like Fourier Transforms and operators related to
linear algebra and random numbers. The ability to manipulate multidimensional
arrays assumes importance as they are used in various algorithms including
those of machine learning.
The ndarray is the most important object in NumPy, as it helps to create an array.
These arrays are zero based indexed. The elements of the array are assigned
consecutive memory locations. These elements are then arranged in the row
major style (as in the case of C) or column major style (as in the case of
FORTRAN). In Python a basic array can be created using the numpy.array
function. The function takes the following arguments:
Object: It returns an array or nested sequence

dtype: The data type of the array is represented by dtype
copy: The object is copied if the value of this argument is true
order: An array can be row major (C) or column major (F) or any (A)
subok: Returned array forced is to be a base class array
ndmin: This argument specifies the minimum number of dimensions
In order to understand the concept, let us have a look at some simple examples.
In the code that follows, the NumPy package is imported as 'np'. The array
function of np is then used to create an array ([7, 14, 21, 28, 35]). Finally, the
value of the array1 is displayed:
>>>import numpy as np
>>>array1 = np.array([7, 14, 21, 28, 35])
>>>array1
array([ 7, 14, 21, 28, 35])
Let us have a look at another example. In the following code, the NumPy
package is imported as 'np'. The array function of np is then used to create a
two dimensional array ([1, 2, 3], [4, 5, 6], [7, 8, 9]). That is the first row
of the array is [1, 2, 3]; the second row is [4, 5, 6] and the third row is [7, 8,
9]. Finally, the value of the array2 variable is displayed:
>>>array2=np.array([[1,2,3],[4,5,6],[7,8,9]])
>>>array2
array([[1, 2, 3],
[4, 5, 6],
[7, 8, 9]])
An array of string type can be created by specifying the datatype (dtype = str) in
the array function of np. The procedure has been depicted in the following
snippet:
>>>array2=np.array([[1,2,3],[4,5,6],[7,8,9]], dtype=str)
>>>array2
array([['1', '2', '3'],
['4', '5', '6'],
['7', '8', '9']],
dtype='<U1')
One of the most obvious questions is why the need to use these arrays in place of
those provided by Python? The first reason is the wide variety of data types
provided by NumPy. The variety of data types provided by the package makes
the task at hand easy.
Some of the important data types provided by NumPy are as follows:
bool_: Boolean
int_: default integer type
intc: Same as that of ‘C’
NumPy also provides int8, int16, int32, and int64 data types. The first (int8)
uses 8 bits and hence can store (-27) to (27-1) (that is -128 to +127). int16
takes 16 bits and hence can store values from -32,768 to +32,767. Likewise
int32 and int64 take 32 and 64 bits. The corresponding unsigned integers are
unit8, uint16, uint32, and uint64. The difference between int and uint is that
in the former, one bit is reserved for sign and in the latter all the bits are used for
storing the numbers, and hence the unsigned integers can store larger values. For
example unit8 can store values from 0 to 255. The data types for floating point
numbers are as follow:
float_
float16: It has 10 bit mantissa
float32: It has 23 bit mantissa
flaot64: It has 52 bit mantissa
The data types for complex numbers are as follows:

complex_
complex64
complex128
Here it may be stated that the data type object describes the interpretation of
fixed blocks of memory. It has information about:
Type of data
Size
Order
The shape and data type in case of a subarray
18.3 FUNCTIONS FOR GENERATING SEQUENCES

Having gone through the basics of NumPy arrays, let us visit some of the
remarkable functions which help when creating useful arrays.
18.3.1 arange()
The arange function helps to print a sequence, having some initial value
(start), some final value (stop), the difference between the consecutive terms
(step), and the datatype (dtype). The syntax of the function is as follows. The
description of each of the parameters follows:
numpy.arange(start, stop, step, dtype)
Start: The starting value of the sequence

Stop: The value up to which the sequence is generated (not inducing the
value itself)
Step: The difference between the consecutive values
dtype: The data type
The above function has been exemplified in the following code which generates
an arithmetic progression having first term 3, the last term 23 (less than 25), and
the difference between the two consecutive terms 2. The data type of the
elements is 'int'.
>>a=np.arange(3,25,2, int)
>>a
array([3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23])
arange can also take a single argument. Writing np.arange(6) will generate a
sequence having first value 0, the difference between the consecutive terms as 1,
and the last term 5, that is:
>>b=np.arange(6)
>>b
array([0, 1, 2, 3, 4, 5)]
In the arange function one can also specify the data type along with the above,
in order to change the default data type of the elements.
>>c=np.arange(6, dtype=float)
>>c
array([0., 1., 2., 3., 4., 5.])
18.3.2 linspace()
The linspace function divides the given range into a specified number of
segregations and returns the sequence so formed. The function takes the
following parameters:
start: The first value of the sequence

stop: The last value (included until endpoint = False)
num: The number of items
endpoint: If endpoint is False then the “stop” value is not included in the
sequence
retstep: If this is True, the step size is returned
dtype: The data type of the elements of the sequence; if not specified then
the data type is inferred by the start and the stop values
Example: In the sequence generated by the following code the first number of
the sequence is 1 and the last value is 27. The number of elements in the
sequence is 11.
>>d=np.linspace(11, 227, 11)
>>d
array([11., 12.6, 14.2, 15.8, 17.4, 19., 20.6, 22.2, 23.8,
25.4, 27.])
If the value of the “endpoint” argument is false, the last value (in this case, 27) is
not included.
>>e=np.linspace(11, 227, 11, endpoint=False)
>>e
array([11., 12.45454545, 13.90909091, 15.36363636,
16.01010102, 18.27272727, 19.72727273, 21.10101010,
22.63636364, 24.09090909, 25.54545455])
The value of the step can be viewed by assigning “True” to the retstep;
argument. For example, in the sequence generated by dividing the range 11-26 in
12 parts, the gap (the last argument of the result) is 1.45454545454546.
>>f=np.linspace(11, 227, 11, endpoint=False, retstep=True)
>>f
(array([11., 12.45454545, 13.90909091, 15.36363636,
16.01010102, 18.27272727, 19.72727273, 21.10101010,
22.63636364, 24.09090909, 25.54545455]), 1.4545454545454546)
18.3.3 logspace()
The logspace function generates a sequence which is equally spaced in the
logspace. That is the elements of the sequence will be between basestart and
basestop “,” where the value of the base is provided in the argument. The default
value of base is 10. The dtype signifies the data type of the elements. Like in the
case of the previous function, endpoint= False would exclude basestop NumPy.
logspace (start, stop, num, endpoint, base, dtype)
18.4 AGGREGATE FUNCTIONS

The numpy module contains many aggregate functions. The various functions
along with a brief explanation are as follows:
numpy.sum This finds the sum of the elements of the argument (e.g. a list
or an array)
numpy.prod This finds the product of the elements of the argument (e.g. a
list or an array)
numpy.mean This finds the mean of the elements of the argument (e.g. a
list or an array)
numpy.std This finds the standard deviation of the elements of the
argument
numpy.var This finds the variance of the elements of the argument
numpy.max This finds the maximum element of the argument. In the case
of a list or a 1D array, the maximum element will be displayed. However, in
the case of a 2D array the axis along which the maximum element is desired
can also be mentioned. Here, axis=0 indicates columns and axis=1 indicates
rows.
numpy.min This finds the minimum element of the argument. In the case of
a list or a 1D array, the minimum element will be displayed. However, in
the case of a 2D array the axis along which the minimum element is desired
can also be mentioned. Here, axis=0 indicates columns and axis=1 indicates
rows.
numpy.argmin This finds the position (index) of the maximum element
numpy.argmax This finds the position of the minimum element
numpy.median This finds the median of the elements of the argument
numpy.percentile This finds the percentile of the elements of the argument.
The percentile (25 etc. ) is the second argument.
numpy.any This finds if any element of the given argument is there
numpy.all This finds if all the arguments of the given argument are there
The following code exemplifies the above functions by generating a set of 50

values. The values are between 0 and 100 (the np.random.random(50) has been
multiplied by 100). The maximum, minimum, argument of the maximum,
argument of the minimum, average, median, standard deviation, variance, 25th
percentile, and 75th percentile of the elements have been found using the above
functions.
import numpy as np
Values1=100*(np.random.random(50))
Values1
Output
Output
The data can be visualized using the matplotlib module discussed in the next
chapter. The following code plots the histogram of the above data.
plt.hist(Values1)
plt.title('Values Generated')
plt.xlabel('Values')
plt.ylabel('Number')
plt.show()
The above functions can also be applied on 2 dimensional arrays, as stated

earlier. The following code exemplifies the above functions by generating a 2
dimensional array having three rows and three columns. The values are between
0 and 1. The maximum, minimum, argument of the maximum, argument of the
minimum, average, median, standard deviation, variance, 25th percentile, and
75th percentile of the elements have been found using the above functions.
Max=np.max(B)
Max_Index=np.argmax(B)
Min=np.min(B)
Min_Index=np.argmin(B)
Sum=np.sum(B)
Prod=np.prod(B)
Mean=np.mean(B)
SD=np.std(B)
Variance=np.var(B)
Med=np.median(B)
Per25=np.percentile(B,25)
Per75=np.percentile(B,75)
print("Max\t:",Max,"\nIndex\t:",Max_Index,"\nMin\t:",Min,
"\nIndex\t:",Min_Index,"\nAverage\t:",Mean,"\nStad
Deviation\t:",SD,"\nVariance\t:",Variance,"\nMedian\
t:",Med,"\nPercentile 25\t:",Per25,"\nPercentile
75\t:",Per75)
Output
Max : 0.960839776764
Index : 3
Min : 0.0795636675955
Index : 1
Average : 0.533217195093
Stad Deviation : 0.345312622379
Variance : 0.119240807174
Median : 0.653392931307
Percentile 25 : 0.180674642779
Percentile 75 : 0.884705912786
Let us now apply the above functions to solving part of a popular problem.
Travelling salesman problem is a popular problem, which finds the shortest
circuit that covers all the vertices from a given graph. There are many ways to
solve the problem. Well, most of them require large amounts of time. However, a
method that uses dynamic programming finds the reduced matrix of the given
matrix. The reduced matrix can be found as follows. First of all, the minimum
element from all the columns is found. The minimum of a column is then
subtracted from the elements of that particular column. The result of this step
will be a matrix having at least one zero in each column. Likewise the minimum
element of each row is found. This element is then subtracted from each element
of the row. The result of this step will be a matrix having at least one zero in
each row. The resultant matrix will be henceforth referred to as the reduced
matrix. In Python this task can be accomplished easily. The first part of the
process has been shown in the following snippet. The reader is expected to
repeat the task for each row as well.
Min_Indeces_Col=np.argmin(B,axis=0)
Min_Col=np.min(B,axis=0)
print(Min_Col)
print(Min_Indeces_Col)
Output:
[ 0.18067464 0.07956367 0.14794398]
[0 0 2]
print(B-Min_Col)
print(np.sum(Min_Col))
Output
18.5 BROADCASTING
If two arrays of same dimensions are added together, an element of the resultant
array is the sum of the corresponding elements of the two arrays. For example, if
Likewise, if
It may be stated here that the above result is obvious. However, Python applies
the operations element by element not only in addition or subtraction, but also in
the case of multiplication and division. The following snippet shows the results
when two row arrays are added, subtracted, multiplied, or divided. (Note that in
each case the resultant array is obtained by applying the operation in the
corresponding elements).
import numpy as np
A=np.array([1,2,3])
B=np.array([7,8,9])
Sum=A+B
Diff=A-B
Prod=A*B
Div=A/B
print("Sum\t:",Sum,"\nDifference\t:",Diff,"\nProduct\
t:",Prod,"Div\t:",Div)
Output
Sum : [ 8 10 12]
Difference : [-6 -6 -6]
Product : [ 7 16 27] Div : [ 0.14285714 0.25 0.33333333]
The above examples give expected results as the dimensions of the arrays on
which the operations were applied had appropriate dimensions. What if two
arrays of different dimensions are added together? Python has a way to deal with
such situations. This is called broadcasting. It may be stated here that
broadcasting does not always work. It works in the situations discussed as
follows. For example, if a row array (say, [1, 2, 3]) is added to a column array
(say [[7], [8], [9]]), then the following procedure is employed:
That is the row matrix is converted to a two dimensional matrix by copying the
elements of the first row to all other rows. Here the number of rows would be
equal to the number of elements in the column matrix. Likewise the column
matrix is also converted into a two dimensional matrix by copying the elements
of the column to all other columns. Here, the number of columns would be equal
to the number of elements in the row matrix. Likewise the results of subtraction,
multiplication, and division can be evaluated. The program follows:
C=B[:,np.newaxis]
C
array([[7],
[8],
[9]])
Sum=A+C
Diff=A-C
Prod=A*C
Div=A/C
print("Sum\t:\n",Sum,"\nDifference\t:\n",Diff,"\nProduct\t:\
n",Prod,"\nDiv\t:\n",Div)
Output
Sum :
[[ 8 9 10]
[ 9 10 11]
[10 11 12]]
Difference :
[[-6 -5 -4]
[-7 -6 -5]
[-8 -7 -6]]
Product :
[[ 7 14 21]
[ 8 16 24]
[ 9 18 27]]
Div :
[[ 0.14285714 0.28571429 0.42857143]
[ 0.125 0.25 0.375 ]
[ 0.11111111 0.22222222 0.33333333]]
If a row array is added to a two dimensional array, the addition is carried out by
converting the first array into a two dimensional array by copying the elements
of the row to all other rows, the number of rows being the same as that in the
second array. That is,
Likewise the subtraction, multiplication, and division can be carried out. The
following snippet shows the results of these operations:
D=[[1,2,3],[4,5,6],[7,8,9]]
Sum=D+A
Diff=D-A
Prod=D*A
Div=D/A
print("Sum\t:\n",Sum,"\nDifference\t:\n",Diff,"\nProduct\t:\
n",Prod,"Div\t:\n",Div)
Output
Sum :
[[ 2 4 6]
[ 5 7 9]
[ 8 10 12]]
Difference :
[[0 0 0]
[3 3 3]
[6 6 6]]
Product :
[[ 1 4 9]
[ 4 10 18]
[ 7 16 27]]
Div :
[[ 1. 1. 1. ]
[ 4. 2.5 2. ]
[ 7. 4. 3. ]]
If a column array is added to a two dimensional array, the addition is carried out
by converting the first array into a two dimensional array by copying the
elements of the columns to all other columns, the number of columns being the
same as that in the second array. The results of addition, subtraction,
multiplication, and division of B and D are as follows (B and D above).
D=[[1,2,3],[4,5,6],[7,8,9]]
Sum=D+B
Diff=D-B
Prod=D*B
Div=D/B
print("Sum\t:\n",Sum,"\nDifference\t:\n",Diff,"\nProduct\
t:\n",Prod,"Div\t:\n",Div)
Output
Sum :
[[ 8 10 12]
[11 13 15]
[14 16 18]]
Difference :
[[-6 -6 -6]
[-3 -3 -3]
[ 0 0 0]]
Product :
[[ 7 16 27]
[28 40 54]
[49 64 81]] Div :
[[ 0.14285714 0.25 0.33333333]
[ 0.57142857 0.625 0.66666667]
[ 1. 1. 1. ]]
The above discussion can be summarized as follows:
If two arrays have different dimensions, the one with lesser dimensions is
padded with the one on the leading side.
If one of the arrays is either row (or column matrix) the elements of the row
(or column) are copied to all other rows (or columns) and then the operation
is applied.
If the dimensions do not match, an error is raised.
18.6 STRUCTURED ARRAYS

The structured arrays help us to create an array having a structure as its element.
This not only helps in maintaining the information easily but also helps with
easy access and manipulation. In order to understand the concept, let us take an
example. Suppose that you have been asked to store the data of employees: their
names, ages, and salaries. Now, you make three different arrays: name, age, and
salary, having the following data:
name=['Harsh','Naved','Aman','Lovish']
age=[100,70,24,18]
salary=[75500.00,65500.00,55500.00,45500.00]
Since it is difficult to handle individual arrays, we create a structured array

containing the above information. In order to do that we can create an array in
which each element is a tuple containing three values: name, which is a Unicode
string of size 10; age, which is an integer of 4 bytes and salary, which is float (8
bytes), using the following statement:
data=np.zeros(4,dtype={'names':('name','age','salary'),'
formats':('U10','i4','f8')})
This is followed by associating the attributes with the
above arrays.
data['name']=name
data['age']=age
data['salary']=salary
The following code depicts an example of creating a
structured array.
import numpy as np
name=['Harsh','Naved','Aman','Lovish']
age=[100,70,24,18]
salary=[75500.00,65500.00,55500.00,45500.00]
data=np.zeros(4,dtype={'names':('name','age','salary'),
'formats':('U10','i4','f8')})
data['name']=name
data['age']=age
data['salary']=salary
print(data)
Output
[('Harsh', 100, 75500.0) ('Naved', 70, 65500.0) ('Aman', 24,
55500.0)
('Lovish', 18, 45500.0)]
Once a structured array is created, the data can be accessed in the usual way. For
example, to display the value of a particular attribute, we can simply mention the
name of the attribute in single quotes inside the square brackets.
data['name']
Output
array(['Harsh', 'Naved', 'Aman', 'Lovish'], dtype='<U10')
data['age']
array([100, 70, 24, 18])
data['salary']
array([ 75500., 65500., 55500., 45500.])
data[1]
('Naved', 70, 65500.0)
These structured arrays can be used to access more complicated information

from the data as well. For example, to see the name of the last employee of the
list we can write:
data[-1]['name']
Output
'Lovish'
18.7 CONCLUSION
The NumPy package helps a programmer to deal with the multidimensional arrays
in the most sophisticated way. The package allows us to create arrays of all
types. The package provides a wide range of aggregate functions to deal with the
elements and analyze the data. Here it may be noted that unlike conventional
matrices, the elements of an array can also be operated upon in a unique fashion,
discussed in the last but one section of this chapter. The chapter also sheds light
onto the importance and usage of multidimensional arrays. The next chapter uses
the matplotlib package to visualize the data and analyze the results. The reader
is expected to go through the next chapter and deal with the problems given in
the appendix of this book.
POINTS TO REMEMBER
NumPy is a Python package, which stands for Numerical Python.
NumPy contains multidimensional array objects and the routines which
process these arrays.
The ndarray helps in creating an array. These arrays are zero based
indexed.
The elements of an array are assigned consecutive memory locations.
The elements of an array can be arranged in the row major style (as in the
case of “C”) or column major style (as in the case of Fortran).
A basic array can be created using the numpy.array function.
The linspace(), logspace() and arange() functions help to create arrays
which have specific sequences.
The structured arrays help us to create an array having a structure as its
element.
Broadcasting helps in applying arithmetic operations to arrays that have
different dimensions
Broadcasting does not always work
EXERCISES
1. In which of the following cases is dynamic typing redundant?

(a) When the type of the variable would not change
(b) When an object needs to hold more than one type of data
(c) Both
2. NumPy is
(a) Numeric Python
(b) Numeric
(c) Number Python
3. NumPy primarily deals with

(a) Multidimensional arrays
(b) Graphics
(c) Animations
4. Which of the following was the predecessor of NumPy?

(a) Numeric
(b) Number
(c) MATLAB
5. Which package provides MATLAB-like capabilities to Python?

(a) NumPy
(b) re
(c) math
6. Consider the following code
A = np.array([4, 2, 1])
B = np.array([3, 9, 27])
C = B[:np.newaxis]
sum = A+C
Which of the following options is sum?
(a) [[7, 5, 4], [13, 11, 10], [31, 29, 28]]
(b) The arrays cannot be added
(c) [7, 11, 28]
(d) [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
7. In the above question, if diff = A–C, which of the following options is diff?
(a) [[1, –1, 2], [–5, –7, –8], [–23, –25, –26]]
(c) [1, –7, –26]
(d) [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
8. In the above question, if prod = A * C, which of the following options is

prod?
(a) [[12, 6, 1], [36, 18, 9], [108, 54, 27]]
(c) [12, 18, 27]
(d) [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
9. If A = [3, 4, 5] and B = [[0, 1, 2],[3, 4, 5], [6, 7,8]], find A–B

(a) The arrays cannot be subtracted
(b) [[3, 3, 3,], [0,0,0], [-3, -3, -3]]
(c) [0, 0, 0]
(d) [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
10. In Python an array can have tuple as its element. Which of the following is
true with respect to this statement?
(a) Such arrays are called structured arrays
(b) An array must have a single data type
(c) This is possible in the case of lists not arrays
11. In Python, which of the following is true with respect to broadcasting?

(a) It is used when the dimensions of the arrays do not match
(b) It is not always possible
(c) Both
12. In Python, which function can be used to generate a sequence?

(a) arange
(b) linspace
(c) logspace
13. The aggregate function in NumPy can be used to find

(a) Mean
(b) Median
(c) Maximum
(d) Minimum
14. Which of the following is true for NumPy?

(a) It is a package which deals with multidimensional arrays
(b) It is useful for carrying out statistical analysis
(c) It works well with MatPlotLib
15. One can create a histogram of the given data using

(a) re
(b) Matplotlib
(c) Both
THEORY
1. Explain the importance of the NumPy module.
2. Explain how a one dimensional array containing zeros is generated using

NumPy. Also explain the creation of a 2 dimensional array.
3. Write a short note on aggregate functions in NumPy. State the functions

used to find the following:
Mean
Maximum
Minimum
Standard deviation
Median
Percentile
4. Explain how an array can be generated in NumPy. Explain the syntax and
usage of the following functions:
Linspace
Logspace
Arange
5. What is broadcasting in NumPy? Explain the rules of broadcasting.
6. What is a structured array? How is a structured array created in Python?
7. Explain the concept of dynamic typing in Python.
8. Explain the need to have the data type of a variable.
9. Explain the procedure of plotting a histogram of a given data.

10. Explain the use of NumPy in generating arithmetic progression and geometric
progression.
APPLICATIONS/ NUMERICAL
1. Ask the user to enter the value of n. Now create an array, a, containing
integers from 0 to (n-1).
2. Create another array, b, from the above array containing all the even
numbers of the original array.
3. Create an array, c, from an array containing all the odd numbers of the
original array.
4. Now add b and c and divide each element of the resultant array by 2. Check
if the result is same as a.
5. Create a one-dimensional array containing 500 random numbers.
6. Find the mean, standard deviation, median, 25th percentile, 75th percentile of
the numbers.
7. Create a histogram of the above data with 10 bins.
8. Implement linear search. Also use the requisite method of NumPy and
compare the running time of both.
9. Sort the elements of the array. Also use the requisite method of NumPy and
compare the running time of both.
10. Create an array of 500 random numbers. Find the product of the numbers
using loops and by using the functions of numpy and compare the running
time by the two methods.
11. From the above array find the maximum element by the following methods:
(a) Using the maximum function of NumPy
(b) Using loops in O(n) time
(c) Using divide and conquer in O(log n) time
12. Create a two-dimensional array having n rows and m columns containing:
(a) All ones
(b) All zeros
(c) 1’s at the diagonal
(d) 0-(m-1) at the diagonal
(e) Random numbers
13. Create a two-dimensional array having 7 rows and 7 columns such that an
element aij (element at the ith row and the jth column) is (i+j)2.
14. Find the sum of elements at the diagonals.

(Refer to Q. 13)
15. Find the maximum element in each row.

(Refer to Q. 13)
16. Find the maximum element in each column.

(Refer to Q. 13)
17. Sort the elements of each row.

(Refer to Q. 13)
18. Sort the elements of each column.

(Refer to Q. 13)
19. From this array create an array having elements of alternate rows and
alternate columns.
(Refer to Q. 13)
20. Create another array, from the original array, having elements from the 4th
row onwards.
(Refer to Q. 13)
21. Create three arrays having the names, age, and roll numbers of students of
an institute.
22. Create a structured array having a tuple as its element, containing name, age
and roll number. Now carry out the following tasks vis-à-vis the structured
array:
(a) Display the names of all the students
(b) Display the ages of all the students
(c) Display the roll numbers of all the students
(d) Display the name of the last student
(e) Display the name of the eldest student
(f) Display the name of the youngest student
(g) Display the names and roll numbers of all the students having an age
entered by the user.
CHAPTER
19
INTRODUCTION TO MATPLOTLIB
• Understand the importance of MATPLOTLIB
• Create the plots for lines, curves, etc.
• Understand subplotting
• Create three dimensional plots using MATPLOTLIB
19.1 INTRODUCTION
In the previous chapter, the methods and procedures used to deal with the data
were discussed. The chapter introduced numpy which helps to accomplish
numerous tasks. It is important to be able to visualize the data as well.
Visualization gives an insight of the results and may help to uncover the
underlying patterns. Matplotlib is a package that helps to plot various types of
graphs and visualize the data. This chapter primarily discusses the pyplot
package of matplotlib.
The pyplot collection of the matplotlib provides a set of functions which help
programmers to perform various tasks associated with plotting. These functions
provide MATLAB like capabilities to Python programmers. pyplot provides
functions to plot a figure, create a plotting area, assign labels, etc. The pyplot
keeps an account of the current figure and the area and hence can direct the
functions to the requisite axes. The following sections discuss some of the most
important functions in pyplot and present some interesting examples.
The chapter has been organized as follows. Section 19.2 introduces basic
plotting, Section 19.3 discusses sub-plotting and Section 19.4 presents 3D plots.
The last section concludes the chapter.
19.2 THE PLOT FUNCTION
We start by plotting the values of a list. In order to generate a basic plot, a list L,
having n values (index: 0 to (n-1)), can be passed to the plot function. In the
generated plot, the x-axis will have values from 0 to (n-1) and the Y-axis will
have the values in the given list. The xlabel function associates a label with the
X-axis. The argument of the xlabel function is a string. The ylabel function
associates a label with the Y-axis. The show function displays the figure. One can
also save the figure using the savefig function. The savefig function takes two
arguments: the figure to be plotted and the dpi. In the following example, the list
L =[1, 4, 8, 10] is passed to the plot function. The string "X Axis" is passed to
the xlabel function and "Y Axis" is passed to the ylabel function. Note that in
the figure, the X axis has values from 0 to 3 (the indices of the values of the list)
and the Y axis ranges from 0 to 10 (from 0 to the maximum value of the list).
This can be changed using the axis function, which takes a list as an argument.
The function has the following arguments:
xmin
ymin
xmax
ymax
The figure is saved as a png file using the savefig function. The function takes
two arguments: the name of the figure and the dpi. Figure 19.1 shows the output
of the program.
plt.plot([1,4,8,10])
plt.xlabel("X Axis")
plt.ylabel("Y Axis")
plt.show()
plt.savefig("line.png",dpi=80)
FIGURE 19.1 If a list is passed to the plot function, the values are plotted against the indices of
the list
In the above example, the plot function takes one argument. However, it can also
take two arguments indicating the values of X and Y. The following example
plots y = 2x2 – 3 (Figure 19.2).
for x in [-5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5].
X=[-5,-4,-3,-2,-1,0,1,2,3,4,5]
Y= [2*x*x-3 for x in X]
plt.plot(X,Y)
plt.show()
FIGURE 19.2 The plot function can also take two arguments; the second argument’s values can
be generated using generators or comprehensions
One can even pass a two dimensional array (or list of lists) in plot, in which case
the first element of each row (or list) would be plotted as a separate plot and the
second as a separate plot. Figure 19.3 shows the output of the plot.
X=[[2,3,1],[4,6,3],[6,9,7],[8,10,5],[9,11,7],[10,18,12],
[11,23,14]]
X
plt.plot(X)
plt.show()
FIGURE 19.3 The plot function can take a two-dimensional array as arguments for plotting
multiple lines
The plot function can have an additional argument stating the color of the plot.
The default color is blue and it can be changed as follows. The color argument
of the plot function can be set to a particular value, say 'red' (color = 'red')in
order to generate a plot of the desired color. In the following example, the x-axis
would now span from 0 to 6 and y-axis would span from 0 to 15, owing to the
arguments of the axis function. The output of the following code would be same
as that of the first one, except for the color of the plot and the axis (Figure 19.4).
plt.plot([1,4,8,10], color='red')
plt.axis([0,6,0,15])
plt.show()
FIGURE 19.4 The plot function can also have an argument to set the color of the plot
If one wants to plot only the points and not the lines, an additional argument “o”
can be passed to the plot function, shown as follows. Likewise, the plots
indicated by a square and a triangle can be plotted by giving “s” and “^.” The
code follows and the output of the program has been shown in Figures 19.5 and
19.6.
plt.plot([1,3,4],[7,8,3],'o')
plt.show()
FIGURE 19.5 The plot function can also plot circles using an additional “o” argument
plt.plot([1,3,4],[7,8,3],'o')
plt.plot([1,2,3,4],[2,1,3,5],'s')
plt.plot([1,5,6],[9,10,11],'^')
plt.show()
FIGURE 19.6 The plot function can also plot squares and triangles using additional “s” or “^” as
argument
The following example shows the procedure for plotting the sine and cosine
functions using Matplotlib. The plot, show, and savefig functions have
already been explained in the above examples. In the following code, the X-axis
is divided into 256 parts (from -22/7 to 22/7). The linespace function helps to
accomplish this task. The sine of the X values can be calculated using the sin
function of numpy. Likewise, the cosine can be calculated using the cosine
function. Both the plots are plotted in the same area. The output has been shown
in Figure 19.7.
from matplotlib import pyplot as plt
import numpy as np
X = np.linspace(-np.pi, np.pi, 256, endpoint=True)
C, S = np.sin(X), np.cos(X)
plt.plot(X,C)
plt.plot(X,S)
plt.show()
plt.savefig("SinCos.png", dpi=72)
The color of the plots can be changed by setting the color attribute to the
requisite value. The linestyle can also be set. The pyplot also provides the xlim
and ylim functions for setting the limits of the X- and the Y-axis. The ticks on
the X- and the Y-axis can be set by using the xticks and yticks functions.
These functions take a list containing the values to be displayed on the axes.
plt.figure(figsize=(8, 6), dpi=80)
plt.subplot(1, 1, 1)
X = np.linspace(-np.pi, np.pi, 256, endpoint=True)
C, S = np.cos(X), np.sin(X)
plt.plot(X, C, color="blue", linestyle="-")
plt.plot(X, S, color="red", linestyle="-")
plt.xlim(-4.0, 4.0)
plt.xticks(np.linspace(-4, 4, 9, endpoint=True))
plt.ylim(-1.0, 1.0)
plt.yticks(np.linspace(-1, 1, 5, endpoint=True))
plt.savefig("SinCos.png", dpi=180)
plt.show()
FIGURE 19.7 The sine and cosine function in the same plot
The following code prints the plot for the log function of numpy. The only
difference in the previous and the following example is the use of the log
function of numpy in place of the sin of the cosine function. The output of the
code has been shown in Figure 19.8.
plt.figure(figsize=(8, 6), dpi=80)
plt.subplot(1, 1, 1)
X = np.linspace(0.1,2, 100, endpoint=True)
L = np.log(X)
plt.plot(X, L, color="blue", linestyle="-")
plt.xlim(0, 2)
plt.xticks(np.linspace(0, 2, 100, endpoint=True))
plt.ylim(-1.0, 1.0)
plt.yticks(np.linspace(-1, 1, 21, endpoint=True))
# Save figure using 80 dots per inch
plt.savefig("Log.png", dpi=180)
# Show result on screen
plt.show()
FIGURE 19.8 The log function
The imshow function of the matplotlib shows the image. The following
example plots the magnitude and the phase of complex number between –2π to
2π. The output of the code has been shown in Figure 19.9.
import matplotlib.pyplot as plt
x = np.linspace(-2*np.pi, 2*np.pi, 100)
xx = x + 1j * x[:, np.newaxis]
out = np.exp(xx)
plt.subplot(121)
plt.imshow(np.abs(out),extent=[-2*np.pi, 2*np.pi, -2*np. pi,
2*np.pi])
plt.title('Magnitude of exp(x)')
plt.subplot(122)
plt.imshow(np.angle(out),extent=[-2*np.pi, 2*np.pi, -2*np.
pi, 2*np.pi])
plt.title('Phase (angle) of exp(x)')
plt.show()
FIGURE 19.9 The use of implot to generate the magnitude and phase of a complex number
The plots described above are an excellent way of comparing the functions (say
x2, x3 and x4) by using the power function of numpy. The plot function has a
label argument stating the type of curve. The limits of the x-axis are set from 1
to 20 and that of y-axis is from 0 to 800. The output of the code is shown in
Figure 19.10.
x = np.linspace (0, 10, 50)
y1 = np.power(x, 2)
y2=np.power(x,3)
y3 = np.power(x, 4)
plt.plot(x, y1, label='$x^2$')
plt.xlim((1, 20))
plt.ylim((0, 800))
plt.xlabel('X Axis')
plt.ylabel('Y : Powers')
plt.title('First :$x^2$ Second:$x^3$ Third:$x^4$')
#plt.legend()
plt.savefig("powers.png",dpi=80)
plt.show()
FIGURE 19.10 The plots of x2, x3, and x4
The pyplot function can also be used to plot a histogram. The hist function can
be used to accomplish the task. The hist function has data as an argument. The
color argument of the function associates the color to the histogram. The
optional commutative argument, if True, plots a commutative histogram. The
number of bins indicates the segregations on the x-axis. The following code
exemplifies the function. The output has been shown in Figure 19.11.
data = np.random.randn(100)
f, (ax1, ax2) = plt.subplots(1,2,figsize=(6,3))
ax1.hist(data,bins=10,normed=True,color='blue')
ax2.hist(data,bins=10,normed=True,color='red',cumulative=True)
plt.savefig('histogram.png')
plt.show()
FIGURE 19.11 A histogram can be plotted using the hist function of pyplot
The imshow function plots the data passed as the argument. The matrix generated
by the random function (10 rows and 10 columns) is passed as the argument in
the imgshow function. The colorbar function of the matlpotlib is then invoked
and is finally shown. The following code exemplifies the function and the output
follows (Figure 19.12).
Img = np.random.random((10, 10))
plt.imshow(Img)
plt.colorbar()
plt.savefig('imageplot.png')
plt.show()
FIGURE 19.12 The colorbar function can be used to plot the required plots
The matplotlib can also be used to plot a 3 dimensional function. In order to

plot a 3 dimensional plot, the projection attribute of the subplot function is set
to '3d'. In the example that follows, the X is a set of 500 random numbers, the
Y axis has sin of X and the value of Z is . The plot_wireframe
function takes four arguments: X, Y, Z, and linewidth. The following code
exemplifies the function and the output follows (Figure 19.13)
from mpl_toolkits.mplot3d import axes3d
ax = plt.subplot(111, projection='3d')
X=np.random.random(500)
Y=np.sin(X)
temp=np.absolute(1-(X**2 + Y**2))
#print(temp)
Z=np.sqrt(temp)
#print(Z)
plt.xlim(-1,1)
plt.ylim(-1,1)
ax.plot_wireframe(X, Y, Z, linewidth=0.1)
plt.savefig('wire.png')
plt.show()
FIGURE 19.13 Plotting 3D graphs in MATPLOTLIB
19.3 SUBPLOTS
Suppose you want to compare the results of two different experiments. Having
two figures side by side would greatly ease the task of comparing the results. In
such situations the subplots come to our rescue. The idea is to have smaller
axes in a single figure; in fact any sort of layouts in a given figure. This task can
be accomplished in a number of ways. One of the simplest ways to use the
matplotlib.axes function in matplotlib; -this is taking advantage of the
fact that standard axes can be created using the matplotlib. axes() function.
This function can also take a list of four numbers depicting the left, bottom,
width, and height respectively. The mechanism of creating a subplot is as
follows. The first two coordinates specify the origin of the new axes, considering
the original axes to be of a unit’s length. For example in the following figure
(Figure 19.14), the origin of the subplot is (0.5, 0.5) and the axes are 40% of the
length of the original axes. The arguments of the axes function in this case would
be, therefore, (0.5, 0.5, 0.4, 0.4).
FIGURE 19.14 The axis (original) range from 0.0 to 1.0 (for both X and Y axes). The origin of the
subplot is (0.5, 0.5). The length of the axes of the subplot are 40% of that of the original axes.
The subplots can be generated manually by using the axes function as shown in
the following code. The output of the program follows (Figure 19.15).
import matplotlib.pyplot as plt
import numpy as np
axis1=plt.axes()
axis2=plt.axes([0.5,0.5,0.4,0.4])
plt.show()
FIGURE 19.15 Creating a new subplot manually, using the axes function
To divide the given plot into two plots by creating another horizontal axis (that is
dividing the graph, vertically, into two subplots), the add_axes function can be
used. The function takes four arguments in a list as shown in the following
example. The two subplots are plotted on two axes. Both the axes are created
using the add_axis function. The sin function of numpy is used to plot a sin
curve on the first axis and a cosine curve on the second axis. The values of X
(50 in number, from 0 to 10) are obtained using the linspace function. The
output of the code has been shown in Figure 19.16.
FIGURE 19.16 The first subplot has a sine curve and the second subplot shows a cosine curve
fig=plt.figure()
axis1=fig.add_axes([0.0,0.5,0.8,0.4])
axis2=fig.add_axes([0.0,0.1,0.8,0.4])
X=np.linspace(0,10,50)
axis1.plot(np.sin(X))
axis2.plot(np.cos(X))
plt.show()
The subplot function of matplotlib.pyplot helps to create a subplot with a

grid. The function takes three arguments: the number of rows, the number of
columns and the index. The following example shows nine subplots in a grid.
Note that the arguments of the nine functions are as follows:
(3, 3, 1)
(3, 3, 2)
(3, 3, 3)
(3, 3, 4)
(3, 3, 5)
(3, 3, 6)
(3, 3, 7)
(3, 3, 8)
(3, 3, 9)
The index of the subplot has been displayed in the subplot. Figures 19.17 and
19.18 show the output of the following codes.
FIGURE 19.17 The subplots can be shown individually
X=np.linspace(-np.pi, np.pi, 100)

for i in range(1,10):
plt.subplot(3,3,i)
Y=np.sin(X+i*(np.pi/2))
plt.plot(X,Y)
plt.show()
#plt.text(0,5,0.5,'Subplot\t:', fontsize=18, ha='center'
)
plt.show()
FIGURE 19.18 The subplots can be shown as a single plot also
Note that the distances between various axes of the subplots are not too much
and hence the plot is not that clean. In order to handle such saturations, the
add_subplot function can be used in the pyplot.figure to create axes. The
horizontal and vertical spaces between the various subplots can be specified
using the subplots_adjust function. The following code exemplifies the above
functions and the output follows (Figure 19.19).
fig=plt.figure()
fig.subplots_adjust(hspace=0.4, wspace=0.4)
for i in range(1,10):
axis=fig.add_subplot(3,3,i)
Y=np.sin(X+i*(np.pi/2))
axis.plot(X,Y)
plt.show()
FIGURE 19.19 Using hspace and wspace in plotting subplots
As a matter of fact, matplotlib provides functions to plot the whole grid in one
go too. The pyplot.GridSpec allows more complicated arguments for
sophisticated options. However, both these are beyond the scope of this book.
19.4 3 DIMENSIONAL PLOTTING

Initially MatPlotLib supported only 2 dimensional plotting. However, the
mplot3d toolkit has helped matplotlib evolve and pose a serious challenge to
MATLAB. The toolkit allows the creation of a 3D space by setting the argument
projection (of the axes function) to '3d'. The following codes show examples of
3D plotting. Note that most of the functions used in the following codes have
already been explained. Each code follows the output (Figure 19.20).
from mpl_toolkits import mplot3d
fig = plt.figure()
axis=plt.axes(projection='3d')
plt.show()
The toolkit helps the programmer to plot sophisticated figures by passing the
values of X, Y, and Z to the scatter 3D function of the pyploy.axes. The
function has two more arguments: a and cmap. The cmap argument can be set to,
say, ‘Greens’ or ‘binary’ or any other value (see the following tip). In the
following example, the Z has 200 values from 0 to 20. The sine of these values
constitute X and the square of the values of Z constitute Y. The code that follows
exemplifies the function and the output follows (Figure 19.21).
FIGURE 19.20 A basic 3-dimensional plot
Tip
Possible values of cmap

Accent, Accent_r, Blues, Blues_r, BrBG, BrBG_r, BuGn, BuGn_r, BuPu,
BuPu_r, CMRmap, CMRmap_r, Dark2, Dark2_r, GnBu, GnBu_r, Greens,
Greens_r, Greys, Greys_r, OrRd, OrRd_r, Oranges, Oranges_r, PRGn,
PRGn_r, Paired, Paired_r, Pastel1, Pastel1_r, Pastel2, Pastel2_r,
PiYG, PiYG_r, PuBu, PuBuGn, PuBuGn_r, PuBu_r, PuOr, PuOr_r, PuRd,
PuRd_r, Purples, Purples_r, RdBu, RdBu_r, RdGy, RdGy_r, RdPu,
RdPu_r, RdYlBu, RdYlBu_r, RdYlGn, RdYlGn_r, Reds, Reds_r, Set1,
Set1_r, Set2, Set2_r, Set3, Set3_r, Spectral, Spectral_r, Vega10,
Vega10_r, Vega20, Vega20_r, Vega20b, Vega20b_r, Vega20c, Vega20c_r,
Wistia, Wistia_r, YlGn, YlGnBu, YlGnBu_r, YlGn_r, YlOrBr, YlOrBr_r,
YlOrRd, YlOrRd_r, afmhot, afmhot_r, autumn, autumn_r, binary,
binary_r, bone, bone_r, brg, brg_r, bwr, bwr_r, cool, cool_r,
coolwarm, coolwarm_r, copper, copper_r, cubehelix, cubehelix_r,
flag, flag_r, gist_earth, gist_earth_r, gist_gray, gist_gray_r,
gist_heat, gist_heat_r, gist_ncar, gist_ncar_r, gist_rainbow,
gist_rainbow_r, gist_stern, gist_stern_r, gist_yarg, gist_yarg_r,
gnuplot, gnuplot2, gnuplot2_r, gnuplot_r, gray, gray_r, hot, hot_r,
hsv, hsv_r, inferno, inferno_r, jet, jet_r, magma, magma_r,
nipy_spectral, nipy_spectral_r, ocean, ocean_r, pink, pink_r,
plasma, plasma_r, prism, prism_r, rainbow, rainbow_r, seismic,
seismic_r, spectral, spectral_r, spring, spring_r, summer,
summer_r, terrain, terrain_r, viridis, viridis_r, winter, winter_r
Z=np.linspace(0,20,200)
X=np.sin(Z)
Y=[z*z for z in Z]
axis.scatter3D(X, Y, Z, c=Z,cmap='Greens');
plt.show()
FIGURE 19.21 The X-axis is sine Z and Y-axis is Z2
In the illustration that follows, the X and the Y have 20 values from -5 to 5. That
is,
Y is the same as X. Z contains the square root of the sum of squares of X and Y,
that is,
The label of the X axis can be set using the set_xlable function. Likewise, the
set_ylabel and set_zlabel function sets the labels for the Y and the Z axis.
The output of the plot has been shown in Figure 19.22.
X=np.linspace(-5,5,20)
Y=np.linspace(-5,5,20)
X,Y = np.meshgrid(X,Y)
Z=np.sqrt(X*X + Y*Y)
fig =plt.figure()
axis.contour3D(X,Y,Z,50,cmap='Greens')
axis.set_xlabel('X axis')
axis.set_ylabel('Y axis')
axis.set_zlabel('Z axis')
plt.show()
FIGURE 19.22 Plot of
Note that the whole figure can be rotated by any angle along the XY plane and
any angle counter clockwise about the Z axis. The view_init function of the
pyplot. axes helps to accomplish this task. The function takes two arguments:
the angle about the XY plane and that (counterclockwise) about the Z axis. The
following code is same as the above except for the last but one line, which
rotates the figure. Figure 19.23 shows the output.
fig =plt.figure()
axis.view_init(60,30)
plt.show()
FIGURE 19.23 The rotated plot of Figure 19.22
The wireframe function helps with easy visualization of the plot. The following
illustration plots a wireframe plot of the above example. Likewise the surface
plot of the function can also be plotted using the plot_surface function. On the
basis of the above discussion, the reader is expected to decode the output of the
variants that follow (Figures 19.24 and 19.25).
FIGURE 19.24 The use of wireframe function
FIGURE 19.25 The axis.contour 3D (X,Y,Z,50,cmap=‘Greens’) function plots the above graph
Code
fig =plt.figure()
axis.plot_wireframe(X, Y, Z, color='red')
#axis.view_init(60,30)
plt.show()
Code
fig =plt.figure()
axis.plot_surface(X, Y, Z, rstride=1, cstride=1,
cmap='viridis', edgecolor='none')
#axis.view_init(60,30)
plt.show()
19.5 CONCLUSION
Plotting is important. It is important to be able to visualize the data to analyze
the results and discover patterns. The popularity of MATLAB is partially
attributed to its ability to present mesmerizing graphs. The same can be achieved
in Python using Matplotlib. This chapter introduces plotting using Matplotlib.
The chapter begins with the explanation of basic plots. The lines, sine curve,
cosine curves and so on can be plotted easily using the Matplotlib package.
Matplotlib allows the subplots also. This can be done manually by using the
Matplotlib.axes function. The subplot function can also be used to create
subplots. As stated earlier Matplotlib used to support only 2-D graphics. The
package now supports 3D graphics as well. This has helped the package share
the stage with MATLAB. In order to plot a 3D graph, an additional argument
'projection' is set to '3d'.
GLOSSARY
Pyplot: The pyplot collection of the matplotlib provides a set of functions
which help the programmers to perform various tasks associated with plotting.
POINTS TO REMEMBER
Visualization gives an insight of the results and may help to uncover the
underlying patterns.
A plot created using pyplot can be saved as a png file using the savefig
function.
If one wants to plot only the points and not the lines, then an additional
argument “o” can be passed to the plot function. Likewise, the plots
indicated by a square and a triangle can be plotted by giving “s” and “^”.
The color of the plots can be changed by setting the color attribute to the
requisite value.
The imshow function of the matplotlib shows the image.
The hist function can be used to accomplish the task.
In order to plot a 3 dimensional plot, the projection attribute of the subplot
function is set to "3D."
To divide the given plot into two plots by creating another horizontal axis
(that is dividing the graph, vertically, into two subplots), the add_axes
function can be used.
The horizontal and vertical spaces between the various subplots can be
specified using the subplots_adjust function.
The label of the x-axis can be set using the set_xlable function.
The wireframe function helps with easy visualization of the plot.
EXERCISES
1. Visualization is important because

(a) It gives an insight of the results
(b) It helps to uncover underlying patters
(c) It can be used for reporting
2. Which collection of MATPLOTLIB provides functions to plot functions,

create a plotting area, etc.?
(a) Pyplot
(b) PyPy
(c) PIL
3. Which of the following can be an argument of the plot function of pyplot?

(a) List
(b) String
(c) File
4. X=[-5,-4,-3,-2,-1,0,1,2,3,4,5]
Y= [2*x*x-3 for x in X]
plt.plot(X,Y)
The above code would plot a
(a) Parabola
(b) Ellipse
(c) Hyperbola
5. Which of the following is used to show a curve?

(a) show
(b) display
(c) Both
6. Which function is used to save the plot using pyplot?

(a) savefig
(b) save
(c) saveimg
7. Which of the following cannot be passed as an argument in the plot function

of pyplot?
(a) color
(b) List
(c) Both
8. The plot function can plot circles using which of the following arguments?
(a) o
(b) s
(c) delta
9. Which of the following is true with respect to plots in Pyplot?

(a) Multiple curves can be plotted on the same plot
(b) The plot can be divided into two subplots manually
(c) The subplot function can be used to see various plots in a single plot
10. Which function is used to plot a histogram in pyplot?

(a) hist
(b) mist
(c) jist
THEORY
1. Discuss the importance of visualization. How does visualization help with
analyzing the results?
2. Name a package that helps to plot graphs in Python.
3. Explain how the values of a list can be plotted using pyplot.
4. Explain how multiple lines can be plotted in a single plot.
5. How is subplotting carried out in Python? Explain the splitting of axes and
the subplot function.
6. Explain 3 Dimensional plotting in MATPLOTLIB.
7. How can you change the color of the graph in Pyplot? Also explain the
importance of cmap.
8. Which package, other than pyplot, can be used to plot graphs in Python?
1. Create a list having numbers in arithmetic progression. Ask the user to enter
the first term, the common difference and the number of terms of the
arithmetic progression. Plot the values using the plot function.
2. Create a list having numbers in geometric progression. Ask the user to enter
the first term, the common ratio and the number of terms of the geometric
progression. Plot the values using the plot function.
3. Create a list having numbers in harmonic progression. Ask the user to enter
the values of “a,” “d,” and the number of terms and plot the curve.
4. Plot the point plots of the above curves.
5. Now, plot the above curves in the same plot and compare them.
6. Create subplots showing the curves of questions 1, 2, and 3.
7. In the above question, adjust the horizontal and the vertical distance
between the subplots.
8. There are four types of parabolas: upward, downward, left facing, and right
facing. The equations of the parabolas having vertex at the origin are as
follows:
Upward: x2 = 4ay
Downward: x2 = – 4ay
Right facing: y2 = 4ax
Left facing: y2 = – 4ax
If the values of x range from [10, 10] and the values of y are calculated
using the appropriate equation, (you can use comprehensions to calculate
the values of y), plot the above parabolas on a single plot.
9. Now create a subplot to plot each one of them.
10. Using plotting prove that (sin θ)2 + (cos θ)2 = 1.
11. Plot the curve of tan θ and identify the points of discontinuity.
12. Plot an ellipse having the length of major axis = 10 and the length of minor
axis = 5.
13. Plot a circle having radius = 10 and center at (5,5).
14. Now plot 10 circles having radius 10 and center’s x coordinate varying from
0 to 10. The y coordinate of the center should be 8.
15. Plot a hyperbola. Ask the user to enter a coordinate and display the
coordinate on the plot. The program should display whether the point is
inside the curve or outside it.
20
CHAPTER
INTRODUCTION TO IMAGE
PROCESSING
• Understand the importance of image processing
• Open an image, read it into an object, and write the object in a file
• Understand the concept of clipping
• Extract statistical information from an image
• Perform rotation, translation, and scaling
20.1 INTRODUCTION
The processing of an image and its manipulation has become more important
with time. Image processing is important in not just identifying people and
objects but is used in numerous fields like medicine, mining, networks, etc. The
importance of image processing can be gauged from the fact that it is used in
magnetic resonance imaging (MRI), X-rays, ultrasound, CT (Computer
Tomography) scans, etc. In addition to the above, image processing is used in
industries for surveillance, fingerprint recognition, face recognition,
authentication, signature verification, etc. Image processing techniques are also
used in areas like weather forecasting, remote sensing and in astronomical
studies. Image processing is an involved and intricate task. Interestingly, humans
are better at recognizing faces and objects. The marvelous network of neurons
helps to recognize objects in a fraction of a second. The recognition of objects
and their classification has always fascinated the computing fraternity. As a
matter of fact the fraternity has lately been mimicking the human beings to
develop such classifiers and recognizers.
The past few decades witnessed marked growth in both hardware and software.
This was coupled with the advances in machine learning techniques. The
combination played a pivotal role in the evolution of the field. The image
processing field has since emerged as one of the most important and independent
fields.
In digital image processing images are manipulated. To begin with, one needs to
remove noise from a given image. The anti-aliasing, sharpening, and removing
blurriness are some of the important tasks that are needed to make a given image
better from our perspective (assuming we all are human beings). From the
machine’s point of view, converting the given image to binary, removing
redundant information, and reducing the sampling rate, etc., are important. So
there are two major goals of image processing: improving human perception and
improving machine perception.
In addition to the above, the machine learning techniques have been successfully
applied to identify diseases from various modalities like MRI (magnetic
resonance imaging) and PET (positron emission tomography). Scientists have
also been successful in making predictive models for the above. However, this
requires some serious image processing. The images need to be segmented and
subjected to various feature extraction techniques, etc., to carry out the above
tasks.
Before processing an image, it needs to be stored. An image can be stored as a

two dimensional array or a 3 dimensional array. If an image is stored as a two
dimensional array, then a 512 × 512 size of image would require a 512 × 512
matrix and therefore 262144 units of memory. Now, if each pixel requires an 8
bit integer the total memory requirement will be 2097152 bits. In the case of a 3
dimensional array the value of the color intensity (say the values of R, G, B) will
also be stored for each pixel.
This chapter discusses image processing using standard functions in SciPy and
NumPy, though image processing can also be done in other ways.
The chapter has been organized as follows. Section 20.2 introduces the basics of
image manipulation in Python. The next section discusses the contour function.
Section 20.4 introduces clipping and Section 20.5 discusses basic
transformations like translation, rotation, and scaling. The last section concludes
the chapter.
20.2 OPENING, READING, AND WRITING AN IMAGE

All the homeless utilities have been included in SciPy.misc(). The following
discussion uses an 8 bit greyscale, 512 × 512 image, which can be accessed by
the ascent() function. The reader can also use the face() function for
accomplishing any of the following tasks.
20.2.1 Opening an Image

In order to open and manipulate images, the misc and pyplot packages (the
latter from matplotlib) need to be included. In the code that follows, the object
‘a’ will contain the image generated by the ascent image. The image can be
saved using the imsave function. This function takes two arguments: the name of
the file in which the image will be stored and object (in this case ‘a’). The image
can be shown by the imshow() function of Pyplot, followed by the invocation of
the show() function (Figure 20.1).
fromscipy import misc
importmatplotlib.pyplot as plt
a = misc.ascent()
misc.imsave('ascent.png', a)
plt.imshow(a)
plt.show()
FIGURE 20.1 The ascent image
20.2.2 Reading
In order to read the above image, the imread function can be used. The imread
function takes a png file as an argument and creates a two dimensional array. The
shape and the dtype of the array can be displayed. The output is, as expected,
(512, 512) and the data type is unit 8.
fromscipy import misc
ascent_array = misc.imread('ascent.png')
type(ascent_array)
print(ascent_array.shape)
print(ascent_array.dtype)
Output
(512, 512)
uint8
20.2.3 Writing an Image to a File

The array obtained in the above program can again be converted into a file. The
tofile() function of the array (ascent_array, in the above example) converts
the array into a raw file. The function takes a single argument, which is the name
of the file to be created. The data of the file can be stored in an object
(sayascent_raw). Note that the shape of the array obtained would be 262,144 in
the given example.
Summary
misc.imsave for saving an object as image

misc.imread for reading an image and putting it in a two dimensional array
tofile for converting a two dimensional array to a raw file
fromfile to read data from a .raw file
20.2.4 Displaying an Image

The image can be displayed using the imshow function. The optional ‘cmap’
argument is not used when the image has shape (m, n, 3). The values of the array
in such cases are interpreted as RGB. However for a (m, n) array this argument
depicts the color map of the image. In the following examples the value of cmap
is set to ‘grey’ to display an image in the greyscale and it is set to ‘jet’ to
display a colored image. The enumeration of the values of cmap has been
presented in the chapter. The imshow function can also take the vmin and vmax
argument. Figure 20.2 shows the output of the code. The image can also be
displayed in greyscale (Figure 20.3).
FIGURE 20.2 The ascent image with cmap = jet
FIGURE 20.3 The ascent image in greyscale
importnumpy as np
ascent_array.tofile('ascent.raw')
ascent_raw = np.fromfile('ascent.raw', dtype=np.uint8)
ascent_raw.shape
(262144,)
plt.imshow(a, cmap=plt.cm.jet)
plt.show()
plt.imshow(a, cmap=plt.cm.grey, vmin=30, vmax=200)
plt.show()
The axis can be removed by passing 'off' to the axis function of pyplot.
plt.axis('off')
plt.show()
20.3 THE CONTOUR FUNCTION

The edges of the polygon can be seen using the contour function. The contour
function draws the contour lines and the fcontour function draws the filled
contour lines. The next example shows the subplots created by varying the
second argument of the contour function from [10, 20] to [200, 210]. Figure
20.4 shows the output.
plt.contour(a)
plt.show()
FIGURE 20.4 The contour of ascent
As a matter of fact, the reader can see the output at various arguments of
contour and observe the differences. The following code passes (10, 20) in the
first iteration, 20, 30 in the second iteration and so on. The output is shown in
Fig. 20.5.
for i in range (1,21):
plt.subplot(5,4,i)
plt.contour(a,[10*i,10*i+10])
plt.show()
FIGURE 20.5 The plots for different values of contour
20.4 CLIPPING
A part of the image can be extracted by creating a mask and putting the values in
the region as 0. For example if one wants to extract a circular area from a given
image, the following steps must be followed. First of all a two-dimensional array
corresponding to the image is created. This is followed by creation of the mask.
Note that in the two examples that follow, the first excludes the area above the
circle (Figure 20.6) and the second excludes the area inside the circle (Figure
20.7).
FIGURE 20.6 The region outside the circular region has been excluded
Code 1
ascent1 = misc.ascent()
ascent1[100:120] = 255
lx, ly = ascent1.shape
X, Y = np.ogrid[0:lx, 0:ly]
mask = ((X - lx / 2) ** 2) + ((Y - ly / 2) ** 2) > lx * ly /
4
ascent1[mask] = 0
ascent1[range(400), range(400)] = 255
plt.imshow(ascent1)
plt.show()
Code 2
ascent1[100:120] = 255
lx, ly = ascent1.shape
X, Y = np.ogrid[0:lx, 0:ly]
mask = ((X - lx / 2) ** 2) +((Y - ly / 2) ** 2) < lx * ly /
4
ascent1[mask] = 0
ascent1[range(400), range(400)] = 255
plt.imshow(ascent1)
plt.show()
FIGURE 20.7 The region inside the circular region has been excluded
20.5 STATISTICAL INFORMATION OF AN IMAGE

The statistical information corresponding to a given image can be easily
extracted using Numpy. The max() function returns the maximum value; the min()
function returns the minimum value, the mean function returns the mean and
finally the std function returns the standard deviation. Consider the code that
follows and observe the output:
print(ascent2.mean())
print(ascent2.max(), ascent2.min())
print(ascent2.std())
Output
87.4798736572
255 0
48.7744598771
20.6 BASIC TRANSFORMATION

Consider any involved task in image processing. The task will require us to do
something with the given image: to translate it, rotate it, or scale it, at the least.
Moreover, in animations such transformations are also required. If the position
or the shape of a given figure is to be changed, the transformation discussed in
this section comes to one’s rescue. The transformations, though, defined for a
point are applicable to lines or curves. These are extensively used in the field of
animation, games, etc. We begin our discussion with three basic transformations:
Translation
Rotation
Scaling
20.6.1 Translation
Translation is the movement of a point or a curve. The operation can be
accomplished by changing the x and the y co-ordinates by a fixed amount. Here,
it may be stated that it is not necessary to change the x and the y co-ordinate by
the same value. The translation of a point is easy to comprehend. The translation
of a line has been shown in the following figure (Figure 20.8).
FIGURE 20.8 The line L is translated along a given direction. The translated line is L’
The movement of the line is shown in the above figure. Note that the shape and
size of the line remains the same and only its x and y co-ordinates would change.
That is, for each (x, y)
The above operation can be described in terms of matrices as follows:
In the case of more than one translation, the corresponding matrices are added to
get the translation matrix. That is,
20.6.2 Rotation
Rotation is the movement of a line or a curve from a fixed point in a circular
direction by a certain angle. The fixed point is known as the pivot point and
through this pivot point, a line perpendicular to X-Y axis (in this case) is known
as axis about which the line rotates. Note that the x coordinate and the y
coordinate can be perceived as the projection of a given line along the x and the
y coordinates. That is,
x = r cos a
y = r sin a
After rotation by the angle b, the angle becomes (a + b) and hence the
projections of the final coordinates become:
y′ = r sin (a + b)
= r sin a cos b + r cos a sin b = x cos b + y sin b
x′ = r cos (a + b)
= r cos a cos b – r sin a sin b = x cos b – y sin b
The above operation can be described in terms of matrices as follows:
To apply translation and rotation at the same time, the following transformation
is used:
X′ = tx + (x – tx) cos b – (y – ty) sin b

Y′ = ty + (x + tx) cos b + (y – ty) sin b
M′ – T = CIS(b) * [M – T]
The rotate function, which takes the angle as an argument, can be used to rotate
the figure. The following figures (Figure 20.9 and Figure 20.10) show the
original image and that after rotation.
FIGURE 20.9 Figure before rotation
FIGURE 20.10 Figure after rotation
When you have to rotate the same figure twice, the task can be accomplished by
rotating the figure by an angle α + β where α is the first angle and β is the
second. The premise can be proved as follows:
20.6.3 Scaling
Scaling is changing the size of the curve or a figure by changing its X and Y axis
by some factor. It’s not necessary for the factor to be the same for both the X and
the Y axis. In the formulas that follow, the scale along the X axis is sx and that
along the Y axis is sy.
In the case where translation is followed by scaling, the resultant matrix can be
crafted using the following mathematical formulation.
In case scaling is to be applied twice then the scaling by a factor, which is a

product of the individual scaling factors, can be used. That is,
Figure 20.11 shows the original image and Figure 20.12 shows the image after
scaling.
FIGURE 20.11 The original image

FIGURE 20.12 The image after scaling
20.7 CONCLUSION
The chapter introduces image processing in Python. To start with image
processing, it is important to appreciate that an image can be stored as a multi-
dimensional array. It is important to be able to store an image, to create an image
from an array and to be able to save the image in a file. Also the reader must
understand the concept of transformations before applying transformations to a
given image. The underlying mathematics helps to untangle the intricacies and
helps the programmer to devise a solution in the case of challenging situations. It
is equally important to be able to clip an image and extract statistical information
from it. The chapter discusses all the above and paves the way of the application
of Python in diverse areas requiring image processing. The reader is expected to
have a look at the references given at the end of this chapter for a clearer
understanding.
GLOSSARY
Translation: Translation is the movement of a point or a curve.
Rotation: Rotation is the movement of a line or a curve from a fixed point in a

circular direction by a certain angle.
Scaling: Scaling is changing the size of the curve or a figure by changing its X-
and Y- axis by a factor.
POINTS TO REMEMBER
In order to read an image from an object, the imread function can be used.
The tofile() function of the array converts the array into a raw file.
The edges of the polygon can be seen using the contour function.
A part of the image can be extracted by creating a mask and putting the
values in the region as 0.
The statistical information corresponding to a given image can be extracted
using Numpy. The max() function returns the maximum value; the min()
function returns the minimum value; the mean function returns the mean and
finally the std function returns the standard deviation.
The rotate function, which takes the angle as an argument, can be used to
rotate the figure.
When you have to rotate the same figure twice then the task can be
accomplished by rotating the figure by an angle α + β, where α is the first
angle and is the second angle.
EXERCISES
1. Which of the following is used in processing images in Python?

(a) re
(b) Python Imaging Library
(c) Pillow
2. An image can be stored as a

(a) 2 dimensional array
(b) 3 dimensional array
(c) Both
3. In Python, the two dimensional array in which an image is stored can be

manipulated by
(a) NumPy
(b) re
(c) Both
4. Which of the following constitute basic transformations in image

processing?
(a) Translation
(b) Rotation
(c) Scaling
5. Which of the following functions can be used to rotate an image?

(a) rotate
(b) rot
(c) rota
6. Which of the following can be done using Python Imaging Library?

(a) Clipping
(b) Rotation
(c) Scaling
7. Image processing is used for

(a) Medical diagnosis
(b) Weather forecasting
(c) Animations
8. Which of the following has the rotate function or rotating an image in

Python?
(a) ndimage
(b) ndarray
(c) Both
9. Which of the following can be obtained with the contour function?

(a) Edges of the polygons in a given figure
(b) Statistical information
(c) Both
10. Which of the following can be used for plotting graphs (both 2D and 3D) in
Python?
(a) MATPLOTLIB
(b) NumPy
(c) Scipy
THEORY
1. Explain some of the applications of image processing.
2. How is an image stored in an array? How is this array used to write an

image to a file and then read it later?
3. Explain three basic transformations. Write the mathematical formulations

involving matrices for carrying out the following tasks:
(a) Rotation
(b) Translation
(c) Scaling
4. Prove that if two translation matrices are added, the resultant matrix would
produce the same effect as individual translations.
5. What is pivot point rotation?
6. Explain clipping. How is clipping carried out in Python?
7. Explain the process of extracting the statistical information of an image.
8. What are contours? How are edges of a polygon extracted in Python?

9. Explain the importance of Python Imaging Library. Can the library be used
with NumPy and SciPy?
10. What is the relation between PIL and MATPLOTLIB?
PROGRAMMING
Write a program to read an image and carry out the following tasks.
1. Store the image in an array and extract statistical information out of it.
2. Find the edges of the polygons in the image.
3. Rotate the image by 30 degrees in an counterclockwise direction.
4. Scale the image by a factor of 2 along the x-axis and 3 along the y-axis.
5. Translate the image by tx and ty (entered by the user).
6. Change the image to greyscale.
7. Reduce the image to half the size.
8. Clip the image by making a square of appropriate size.
9. Clip the image by making an ellipse of appropriate size.
10. Store a text in the image.
USEFUL LINKS
1. http://www.scipy-lectures.org/advanced/image_processing/
2. This chapter explains image processing using PIL. Open CV is also an

important package dedicated to image processing. The introduction to Open
CV can be found at http://www.scipy-lectures.org/packages/scikit-image/
3. This chapter explains image processing using PIL. Scikit-image also an

important package dedicated to image processing. The introduction to the
package can be found at: skimage 0.13.0 docs–skimage v0.13.0 docs.
4. The following link gives a brief overview of Pillow:

http://www.bogotobogo.com/python/python_image_processing_with_Pillow_library.php
APPENDIX
A
MULTITHREADING IN PYTHON
A1.1 INTRODUCTION
Modern systems can run many processes simultaneously. In the case of a
multiprocessing system, various processes are allocated different memory space.
However, these processes can share some memory space for purposes like
communication. If a system has a single CPU (central processing unit), it can
distribute the CPU time between these processes. The CPU hops between these
processes and gives a slice of processing time to each process. There are many
ways to carry out this scheduling like First Come First Serve, Shortest Job First,
Round Robin, etc. Multiprocessing helps with achieving efficiency and results in
better utilization of resources. At times a single process may have some portions
that can run independently, provided that the problems like synchronization have
been handled. This gives rise to the idea of multithreading. As per the literature
“Multithreading is the ability of a computer’s operating system to run several
programs or apps at what seems to be the ‘same time’ by a single processing unit
or CPU.”
When a process is expected to process two or more parallel tasks, generally a

thread is implemented. In fact, every process is a single thread of execution and
a thread can give rise to many threads. Using multi-threading, a single program
can perform several of its sub-tasks concurrently. One of the simplest examples
can be the spell check, grammar-check and word count utilities in a word
processor. The word processor here is a main process and the utilities can be the
various threads, which run in parallel. Figure A1.1 shows the differences
between a process and a thread.
FIGURE A1.1 Difference between a process and a thread
Python supports multi-threading. In a multi-threading system, a thread is the

smallest unit of execution. Threads of a process share the same memory space.
Also, the state of a process is shared by a thread. Threads can also share global
variables. If one of the threads is changed by the global variable, others can
access the updated value. These threads are run in parallel by scheduling or by
dividing the CPU time amongst the threads. In the first instance, these threads
might appear to be just an extension of a function. However, a detailed look will
reveal the difference, primarily in the return behavior.
Threads can be segregated into two classes. The threads can be created by the
user or even created by a kernel. The latter are the part of the operating system.
In any case different tasks are being done by different threads, resulting in better
CPU utilization. That is, multi-threading is advantageous vis-a-vis resource
utilization. The computers with many processors can use multi-threading to full
advantage. Moreover, as explained in the above discussion, the time during
which input, etc., is processed can be judiciously used. Figure A1.2 shows the
advantages of multi-threading.
FIGURE A1.2 Advantages of multi-threading

A1.2 THE JAVA THREADING MODULE
The Python threading model is inspired by the Java threading model which has
been explained in this section.
In Java, a thread is created using the thread class or the runnable interface. A
new thread is born when it begins its life cycle. This state is referred to as
“New.” When a thread is born, it becomes “Running.” That is, it becomes ready
to accomplish the task for which it was created. During the execution, there
might be a condition where it waits for some signals, perhaps IO. This state is
referred to as the “waiting state.” It may be stated here that this waiting can also
be timed. Finally, a thread is “terminated.” The states of a thread have been
presented in the following diagram (Figure A1.3).
FIGURE A1.3 States of a thread
A thread can be assigned some priority, based on which the operating system
decides which thread to execute. In Java, the minimum priority that can be
associated with a thread is 1 (MIN_PRIORITY) and the maximum can be 10
(MAX_PRIORITY). As a matter of fact we assign a higher priority to a thread
when it is more important. However, the final decision in this regards is at the
operating system’s discretion.
A1.3 THREADING IN PYTHON

Python has two modules to facilitate the usage of threads: Threads and
Threading, though Threading is used and the former has been deprecated. In
Python 3, the former is called thread. The thread module comes with many
functions to accomplish various tasks. A new thread can be started by using the
start_new_thread method.
The start_new_thread method starts a new thread. The arguments of the

method are function, args, and kwargs. function is the name of the function,
the list of arguments is in args and kwargs is the dictionary of the keyword
arguments.
The following listing shows the usage of the start_new_thread function. The
function, first of all, needs to be imported from the _thread module. Then a
function, say funl is defined. This is followed by any number of invocations of
the threads, as shown. Note that the exception is caught in the function and
hence when a particular thread is invoked, the requisite exception is caught (or
the invocation leads to a normal execution).
def fun(num):
print(‘Hi there\n Number\t:’ + num)
from _thread import start_new_thread try:
start_new_thread(fun,(1,))
start_new_thread(fun,(2,))
except Exception:
print(‘Caught exception’)
The threads can be monitored using counters. A global counter can be created,
which can be incremented on each call to the start_new_thread function and
decremented when the thread exits. As stated earlier, the thread module has
been deprecated so the threading module provides various interfaces for
accomplishing different tasks. The important points regarding the creation of
threads are as follows:
The number of thread objects that are currently active are displayed by the
active_count() function.
The current thread object is returned by the current_thread() function of
the caller’s thread.
The thread identifier can be seen using the get_ident() function. As per
Python.org
“This is a nonzero integer. Its value has no direct meaning; it is intended as
a magic cookie to be used e.g. to index a dictionary of thread-specific data.
Thread identifiers may be recycled when a thread exits and another thread
is created.”
The complete list of all threads running can be viewed by the enumerate()
function. The list contains the main thread and all other threads.
The main thread can be seen using the main_thread(). In order to set the
trace for all threads settrace() can be used.
The size of the stack can be set using the stack_size function.
A thread can have its local data as well. The data specific to a thread can be
stored in an instance of the local class. An instance of the Threading.local can
be created as follows:
data = threading.local
data.x = 1
A1.4 IMPORTANT METHODS OF THE THREAD CLASS

This section briefly introduces some of the most important methods of the
thread class and paves the way for their usage in the requisite code.
_init_() and run()
A thread in Python can be created by instantiating the Thread class. The activity
that a thread would initiate can be represented in two ways. The first is to pass
the object to the init(), that is the constructor of the thread class, in which case
other methods of the thread class should not be overridden. The other way to
initiate the activity is to override the run() method.
start() and isalive()
After creating a thread the requisite activity can be started by calling the start
method. The calling of the start() invokes the separate thread of execution.
The thread initiated would be in the live state. The is_alive() method checks
whether the thread is alive.
join()
If the join() method of another thread is invoked by a thread, the calling thread
is blacked until the thread whose call has been invoked is terminated.
name Attribute
The name attribute of a thread is used to see or change the name of a thread.
A1.5 TYPES OF THREADS

A thread can be a daemon thread, dummy thread or a normal one. A brief
description of the types of threads has been presented as follows.
If a thread is a “daemon thread”, then the program exits if it is left with only
daemon threads. The flag can set through the daemon property. That is they can
be abruptly shunted. The main thread corresponds to the initial thread of control
in Python program. The main thread cannot be a daemon thread.
Then there are dummy threads. These threads are daemonic and are always alive.
These threads cannot be joined with any other thread.
class threading.Thread(group=None, target=None, name=None,

args=(), kwargs={}, *, daemon=None)
The arguments of this constructor are as follows:
group should be none; It is reserved for future extension when a

ThreadGroup class is implemented.
target is the callable object to be invoked by the run() method.
name is the thread name.
args is the argument tuple for the target invocation.
kwargs is a dictionary of keyword arguments for the target invocation.
A1.6 CONCLUSION
This appendix introduces multi-threading. The importance of threading and the
difference between a thread and a process was explained in the first section. It
has been argued that multi-threading is good, both in terms of resource
utilization and efficiency. The Python threading model is largely based on the
Java threading model. The appendix, therefore, gives a brief introduction of the
Java threading model as well. The threading class and its important methods
have been discussed so as to give an idea of how threading is implemented in
Python. The reader is requested to visit the references at the end of the book for
a detailed insight into the topic. It may also be stated here that the purpose of the
appendix is not to discuss each and every aspect of threading but to briefly
introduce it.
EXERCISES
1. What is multitasking?
2. State some conditions where multi-threading cannot be used.
3. Differentiate between a thread and a process.
4. What are the advantages of multi-threading?
5. In case of a uni processing system, what is the use of multi-threading?
6. Name three languages which support multi-threading.
7. Explain the Java threading model.
8. Explain how threads are implemented in Python.
9. Name the libraries which help to implement multi-threading in Python.
10. Name a technique via which thread safely can be implemented.
11. Which exception must be handled for a multi-threading code?
12. What is the function of the join() method?
13. All the threads are waiting for a signal. Which method would notify all of
them?
14. How is a thread stopped in Python (name the method)?
15. What are the various types of threads?
APPENDIX
B
REGULAR EXPRESSIONS
The functions studied in strings and the procedures studied in the third section of
this book helped us to search a pattern from a given text or extract it. However,
searching, finding patterns, and extracting a text are common tasks—so common
that a whole module of Python is dedicated to these tasks. The re module helps
the user to extract or search a pattern from a given text. It is a bit complex and
therefore has not been included in the main text. However, it is important and
has therefore been introduced in this appendix. The aim of this appendix is not to
cover the topic comprehensively and analyze every aspect of the topic. However,
a brief introduction paves the way of the use of the module for simple tasks.
Regular expressions are powerful and can be used to search patterns like e-mail
IDs, phone numbers etc., from a given text. Also, their use in parsing and the
development of compilers is well known.
B1.1 INTRODUCTION
Python provides the users with the re module to deal with the regular
expressions and search requisite texts. Writing a complex regular expression is
an intricate task which requires practice. However, simple expressions can be
written right away. In fact, just by reading this section!
First of all, let us consider the characters in a regular expression. The types of
characters in a regular expression are as follows:
Literal characters
Character class
Modifiers
The literals represent the class that has a single character: upper and lower case,
digits and special characters. The special characters require a backslash (/) to be
placed before the character itself.
Character class has one or more characters in square brackets. The expression
will be matched by the occurrence of any character from the class. As a matter of
fact, one can even mention a range in the class. The following symbols depict
some of the standard matches:
Alphanumeric characters are represented by \w,

The non-alphanumeric characters are represented by \W
The digits are represented by \d
The non-digits are represented by \D
The white spaces are represented by \s
The beginning of a string is represented by ^. The end of a string is
represented by $
In a regular expression the dot (.) depicts a single character
The asterisks (*) represent zero or more occurrences of the preceding
character
The + represents 0 or more occurrence of a string and
? represents 0 or 1 occurrence of the preceding character
In order to match exactly n occurrences of a regular expression, use {n}.
To match n or more occurrences, the {n,} is used.
To match any number of occurrences between n and m {n, m} is used.
A regular expression is generally compiled before being used to accomplish the

said tasks. The comparisons are done only after the given regular expression is
compiled. As a matter of fact, even if a function is accepting an un-compiled
regular expression, then also it is better to use a compiled regular expression as
this greatly reduces the time taken for searching.
A regular expression is written in single quotes, preceded by r. For example, the

name of a website can be any number of alphanumeric characters followed by a
dot and then a domain. This regular expression can, therefore be written as:
sitename= re.compile(r'[\w.]+@[\w.]+')
The presence of a regular expression can be searched in a string using functions

like search and match. These functions can be called as per the expression
whether it is compiled or not:
match: looks for the expression at the beginning of the string

search: looks for the expression everywhere in the string
Each function takes the pos and endpos arguments indicating the beginning and
the end position to be searched in the expression. If the expression is found in
the given text, a match object(s) is returned. If the match is not found then None
is returned. Let us consider the following example:
import re
mailid=re.compile(r'[\w.]+@[\w.]+')
Here, import re is for importing the re module. mailid=re.compile

(r'[\w.]+@[\w.]+') is for creating a regular expression called mailid. Note
that it starts with r and the expression is in ‘_’. The email ID can have any
number of alphanumeric characters followed by a . and then @, which should be
followed by any number of alphanumeric characters and (any number of times).
After being compiled, it can be used to search the requisite string.
text = 'The site of the university if [email protected]'

mailid.search(text)
The output of the above follows:
<_sre.SRE_Match object; span=(29, 44),

match='[email protected]'>
Having seen the basics, let us now dwell into the regular expressions with a
newer and easier perspective.
B1.2 THE SEARCH FUNCTION AND POWERFUL REGULAR

EXPRESSIONS
In a program that uses regular expressions, the re module must be imported. One
of the simplest tasks that can be performed using the module is to use the search
function of the module to find a given pattern. For example, the following code
finds the occurrences of the string “Harsh” in the text from the file “file1.txt”.
The power of re, however, is immense. It can be used to perform rather
complicated tasks like that of finding all the lines that begin with the string
“har”. The following code uses the ^ to accomplish this task:
f=open ('Text1.txt')
i= 1
str=”
for line in f:
The above tasks could also be accomplished by using the existing string
functions. Now, if a little difficult task, say, that of finding all the words that
begin with a “h” followed by 4 symbols followed by a “h” are to be searched
using the re module, first of all the module needs to be imported.
In a regular expression? matches any character. This is one of the most

commonly used special characters used to match an expression. For example, a
H...h would match any word that starts with a “H” and ends with a “h.” That is
“H...h” would match \
Harsh
Haaah
H123h
H@12h
and so on.
The following code implements the above logic.

Output
I am Harsh.
Moreover, we can also specify if the character can repeat. The symbols “*” and
“+” in regular expressions indicate that in place of matching a single instance of
the given character, zero or more instances for a “*” and one or more instances
for a “+.”
Note that there is a “+” after “.” indicating that any number of symbols may be
there between a From and a @
If one wants to extract all the strings that match a regular expression, the
findall() function can be used.
B1.2.1 Extracting data using regular expressions
If we want to extract data from a string in Python we can use the findall(). For
example to find all the email IDs from a given string the expression
'\S+@\S+.com' can be used.
That is, writing
list = findall('\S+@\S+.com',str)
list
would generate a list containing all the email IDs that end with a .com.
Note that \S is a non-whitespace character and \S+ indicates any number of non-
whitespace characters. The @ must follow. After the @ symbol, there can be any
number of non-whitespace characters followed by a .com.
Likewise one can craft a regular expression for finding all .edu emails and so on.
Strictly speaking an email ID starts with a capital or a small letter, followed by

any number of non-whitespace symbols and then a @ sign. The @ follows any
number of non-whitespace characters. The regular expression for e-mail id
would therefore be
[a-zA-Z0-9]\S*@\S*[a-zA-Z]
One of the most common examples that is encountered in literature is that of

substrings that start with a single lowercase letter, uppercase letter, or number
“[a-zA-Z0-9]”, followed by zero or more non-blank characters (“\S*”),
followed by an at-sign, followed by zero or more non-blank characters (“\S*”),
followed by an uppercase or lowercase letter. The code follows:
The regular expressions can also be used to find all the expressions that follow a
particular pattern. For example, if we want to extract patterns that begin with a H
followed by a _ and then a colon followed by any number of digits, then a dot
and then any number of digits, the following re would suffice.
Sample O/P
H-ABCD-lefthanddrive: 0.1234
H-XYZT-righthanddrive: 0.1111
H-NMOP-lefthand: 0.0101
H-KLMN-righthand: 0.0001
H-AAAA-leftdrive: 0.01010
H-C DCD-rightdrive: 1010.90990
We say that we want lines that start with “H-,” followed by zero or more
characters (“.*”), followed by a colon (“:”) and then a space. After the space we
are looking for one or more characters that are either a digit (0-9) or a period
“[0-9.]+”. Note that inside the square brackets, the period matches an actual
period (i.e., it is not a wildcard between the square brackets).
Code
import re
f = open('File1.txt')
for line in f:
line = line.strip()
if re.search('ˆH-\S*: [0-9.]+', line) :
print line
The execution of the above code would throw up the string we are exactly
looking for. Actually the reason for splitting a given text into lines was to spare
the need for parsing and searching at the same time. Parentheses are special
characters in regular expressions. Adding parentheses to a regular expression
helps us to ignore while matching the string.
However, when we use findall() parentheses to indicate that we want the whole
expression to match, we are only interested in extracting a portion of the
substring that matches the regular expression.
import re
f = open('File.txt')
for line in f:
line = line.strip()
pattern = re.findall('ˆH-\S*: ([0-9.]+)', line)
if len(pattern) > 0 :
print(pattern)
The output from this program is as follows:
['0.8475']
['0.0000']
['0.6178']
['0.0000']
['0.6961']
['0.0000']
At times we need to search the characters which are actually special symbols in a
regular expression. In such cases it must be indicated that the characters are
normal and not that with a special meaning. We can indicate that we want to
simply match a character by prefixing that character with a backslash. The
examples follow.
As stated earlier, the purpose of the appendix was to give an introduction of the
topic. Actually, using the regular expressions we can construct strings that are to
be searched/matched with some of the string in the given text.
B1.3 CONCLUSION
At times, big text needs to be searched for a particular pattern. The strings to be
searched can be of various types. For searching for a given pattern, Python
provides us with regular expressions via the re module. For example, in
searching the email IDs from a given document or in finding the names that end
with “a,” regular expressions can be used. Regular expressions, as a matter of
fact, are not just used to find patterns but can also be used to define or modify a
pattern. For example, if you are required to find “all the TV series whose names
begin with a K followed by a k” the task can be accomplished by defining the
requisite re, compiling it and then using it to find the pattern. This appendix
introduces the package and explains its use.
SYMBOLS TO REMEMBER
Some of those special characters and character sequences are as follows:
ˆ: The beginning of the line.
$: The end of the line.
. : Any character (a wildcard).
\s: A whitespace character.

\S: A non-whitespace character (opposite of \s).
*: Indicates to match zero or more of the preceding

character(s).
*?: Indicates to match zero or more of the preceding

character(s) in “non-greedy mode.”
+: Indicates to match one or more of the preceding

character(s).
+?: Indicates to match one or more of the preceding

character(s) in “non-greedy mode.”
[abcde]: The example would match “a”, “b”, “c”, “d”, or “e”, but no
other characters.
[a-z0-9]: Single character that must be a lowercase letter or a digit.
[ˆA-Za-z]: Single character that is anything other than an uppercase or

lowercase letter.
(): Ignored for the purpose of matching, but allows you to

extract a particular subset of the matched string rather than
the whole string when using findall().
\b: Matches the empty string at the start or end of a word.
\B: Matches the empty string, but not at the start or end of a
word.
\d: Matches any decimal digit; equivalent to the set [0-9]
\D: Matches any non-digit character; equivalent to the set [ˆ0-

9]
EXERCISES
THEORY
1. What are regular expressions? How are they useful?
2. Are regular expressions better then string functions?
3. What is the difference between a compiled and a non-compiled regular
expression?
4. Can we use regular expressions to find patterns from a web page?
5. Name three functions used for finding a string from a given text.
PROGRAMMING
1. Take a text of around 2000 words from a news web site. From this text find
(i) The email IDs of all the people who have their mail at the Yahoo server
(ii) The phone number of people living in Jordan (the phone number starts
from 011-)
(iii) The code that begins with an “H,” has two digits after “H” and any
number of characters after the digits.
(iv) All the occurrences of the names that begin with an “H” and ends with an
“h.”
(v) “Ashish” followed by three digits from the given text.
APPENDIX
C
EXERCISES FOR PRACTICE:

PROGRAMMING QUESTIONS
SECTION I: PROCEDURAL PROGRAMMING

Conditional Statements
1. Ask the user to enter a four digit number and check whether the second digit
is one more than the third digit.
2. In the above question if the given condition is false, swap the digits at the
third and the second places and increment the digit at the units’ place by one,
if it is not 9. If the digit at the units place is 9, then do not change the digit.
3. Ask the user to enter a three digit number and find the characters
corresponding to the digits. Create a string out of these characters and find
whether any letter of this string is a capital letter.
4. Ask the user to enter his monthly salary, his house rent (or home loan), his
car, newspaper subscription, the amount he spends on food in a month. Now
find if the amount left is sufficient enough to start a 401K. Note that a 401K
can be started with even $500.
5. Ask the user to enter his total savings. If the savings are above $1,000,000,
then the person need not to pay any tax. Also this person is entitled to get a
subsidy from the Government. If the savings are above $1,000,000 but below
the above specified amount, he is liable to pay 30% of his savings as tax,
plus a surcharge of 2% on the tax. Calculate the total amount paid by the
person as this tax.
6. Ask the user to enter a three digit number and find the largest digit of the
number. Also find the sum of the digits and find if the sum of the digits is the
same as twice the largest digit.
7. Ask the user to enter the marks obtained by a student in 5 subjects. If the
person scores more than 90% in a subject he gets an A+. If the score is less
than 90% but greater than 85%, he gets an A. If the score is greater than
80%, he gets an “A-.” Likewise if the score is greater than 75%, he gets a B+
and B is awarded to a person scoring more than 70%. A person getting more
than 65% but less than 70% gets a “B-” and the one getting more than 60%
but less than 65% gets “C+.” A person getting more than 55% (and less than
60%) gets a “C” and the one getting more than 50% (and less than 55%) gets
a “C-.” Furthermore, for each grade the corresponding CGPA is as follows.
Grade cgPA
A+ 9
A 8
A– 7
B+ 6
B 5
B– 4
C+ 3
C 2
C– 1
Find the average CGPA of the student.

8. Find whether the year entered by the user is a multiple of 7, without using the
mod operator.
9. Find whether the number entered by the user is a multiple of both 5 and 7,
without using the mod operator.
10. Ask the user to enter a string and find the number of occurrences of vowels
in the string.
Looping
11. Ask the user to enter a number and find the number obtained by reversing the
order of the digits.
12. Ask the user to enter a decimal number and find its binary equivalent.
13. Ask the user to enter a decimal number and find its octal equivalent.
14. Ask the user to enter a decimal number and find its hexadecimal equivalent.
15. Ask the user to enter an n-digit number and find the digit which is the
maximum amongst them.
16. Ask the user to enter any number of numbers (he must enter 0 to quit) and
find the maximum number.
17. In the above question find the minimum number.
18. Ask the user to enter n numbers and find their standard deviation and mean.
19. In the above question, find the mean deviation.
20. Write a program to generate the following pattern. (Refer to the link of
cellular automata at the end of this appendix).
Functions
A piece of data is given to you. The data has many features (columns) and the
last column states the class to which it belongs (0 or 1). Each feature’s data can
be segregated into X and Y, where X is the data that belongs to class 0 and Y is
the data that belongs to class 1. The relevance of a particular feature can be
calculated by numerous methods, one of which is the Fisher Discriminate Ratio.
The Fisher Discriminate Ratio of a feature (a column vector) is calculated
using the following formula
Where μx is the mean of the data X, μy is the mean of the data Y. The
standard deviation of X is σx and that of the Y data is σy.
Ask the user to enter the elements of two lists - feature and label.
21. Create a function, segregate, which takes the feature and label as input and
find the vectors X and Y.
22. The calculate_mean function should calculate the mean of the input vector.
23. The calculate standard_deviation function should calculate the standard
deviation of the input vector.
24. The FDR function should calculate the FDR of a feature.
25. Finally write a program that takes 2D data as input and calculate the FDR of
each feature.
The relevance of a particular feature can also be calculated by the
coefficient of correlation.
The coefficient of correlation of a feature (a column vector) is calculated
using the following formula.
26. Create a function, “segregate” which takes the feature and label as input and
find the vectors X and Y.
27. The calculate_mod function should calculate the |X| for the input vector, X.
28. The calculate_dot function should calculate X.Y.
29. The CORR function should calculate the correlation coefficient of a feature.
30. Finally write a program that takes 2D data as input and calculate the
coefficient of correlation of each feature.
File Handling/ Strings
31. Create a file called data and insert data from a text file containing 5 news
articles from a news site.
32. Now open the file and find the words beginning with vowels. Make 5 lists of
words beginning with each vowel.
33. Draw a histogram of the above data.
34. Make the first letter of each word capital and write the words in 5 separate
files.
35. Now, from each file find the words that end with a vowel and place the
words in 5 separate files.
36. Check which of these words begin and end with a vowel.
37. From the original file find the word which is repeated the maximum number
of times.
38. Do the above task for all the words and plot the frequency of each word in a
graph.
39. From the original file find which alphabetic character is used the maximum
number of times.
40. The reader is expected to read about Huffman code from the following link
and encode the file using Huffman code.
https://users.cs.cf.ac.uk/Dave.Marshall/Multimedia/node210.html
41. From the original file, find the string that has the maximum length.
42. From the original file find the string that has “cat” as the substring.
43. From the original file find the strings which are substrings of some other
strings in the file.
44. From the original file find the strings which begin with a capital letter.
45. From the original file, find all the email IDs.
46. Find the email IDs which are on the Yahoo server.
47. Create a regular expression for the land line number in India and find all the
landline numbers from the file.
48. From the above list find the phone numbers which belong to a particular
area.
49. Find the words which are in all five articles.
50. Find the words which end with a consonant and contain a vowel.
SECTION II: OBJECT-ORIENTED PROGRAMMING

Classes and Objects
You are required to develop software for a car wash company. The company
wants software that can store the details of a car and generate invoices. After due
deliberation, it was decided that a class called car with the following members
would be created.
Data members
(a) Registration Number
(b) Model
(c) Make
(d) Year
(e) Name of the owner
Methods
(f) getdata() : Takes data from the users
(g) putdata() : Displays data
(h) ___init() : Initializes members
(i) del : Destructor
(j) capacity :
1. Create a class called car to facilitate the development of the said software.
2. Make two instances of the class and display the data. The first instance
should display the data entered using the putdata() function and the second
should display the data assigned using the init () method.
3. Create an array of cars. Ask the user to enter the data of n cars and display
the data.
4. Find the cars whose registration numbers contain “HR51.”
5. Find the cars which are manufactured by “Maruti.”
6. Find the cars which were manufactured before 2007.
7. Find the car whose owners are named “Harsh.”
8. Find the cars whose owners names begin with “A” and are manufactured
after 2014.
9. Find the cars which have a certain type of engine (entered by the user).
10. Find the car with the maximum engine capacity.
Operator Overloading
11. Create a class called vector, which has three data members
(a) x1: The x component of the vector
(b) y1: The y component of the vector
(c) z1: The z component of the vector
The class should have a method called getdata(), which takes data from
the user; putdata(), which displays the data; init, the constructor.
12. Create a class called vectors and make two instances of vector: v1 and v2.
Display the data of the two objects.
13. The mod of a vector can be defined as follows. If then
Create an array of vectors. Ask the user to enter the data of
n vector and find the vector that has the maximum mod.
14. From the above vectors (Question 13) find the vectors which have the y
component 0.
15. Two vectors v1 and v2 can be subtracted by subtracting the corresponding
components of the two vectors. That is if and
then
Using the above concept, overload the + operator for the class.
16. Two vectors, v1 and v2, can be added by adding the corresponding
components of the two vectors. That is if and
then
Using the above concept, overload the + operator for the class.
17. The dot product of two vectors can be obtained by adding the products
obtained by multiplying the corresponding components of the two vectors.
That is, if and then
Using the above concept, overload the . operator for the class.
18. A hypothetical operation can increment can be defined as follows. If
then
Using the above concept, overload the ++ operator for the class.
19. A hypothetical operation can increment can be defined as follows. If
then
Using the above concept, overload the – – operator for the class.
20. For the vector class, overload the unary (–) operator.
Inheritance
21. Create a class called Book that has the following members.
(a) Name of the book : String
(b) Author(s) : List
(c) Year : Year of publication
(d) ISSN : String
(e) Publisher : Name of the publisher
The class should have getdata(), putdata() and init() as its methods.
22. Create two subclasses: TextBook and ReferenceBook having requisite data
members. Demonstrate the use of overriding in the above hierarchy.
23. Now create three subclasses of the TextBook class, namely SocialScience,
Engineering and Management. Each class should define its version of
getdata() and putdata(). Make instances of these subclasses and call the
method of the derived classes.
24. Create a class called XBook, which is a subclass of both TextBook and
Reference Book and demonstrate how this can lead to ambiguity.
25. Create a class called ABC and craft a class method and an instance method
of the class.
Exception Handling
26. Create a class called array which contains an array and max which is the
maximum number of elements the array can have and methods getdata()
and putdata(), which perform the requisite tasks.
27. Now create a class to raise customized exceptions. The exception should be
raised if the user enters more elopements than the max allowed.
28. If the user enters anything except for integer, an exception should be raised
and the requisite message should be displayed.
29. Now ask the user to enter two indices and divide the numbers at those
positions. If the number at the second position is 0, an exception should be
raised.
30. Ask the user to enter three indices. These three indices contain the values of
“a,” “b,” and “c” of the quadratic equation ax2 + bx + c = 0. Find the
discriminant and the roots of the equation. If the value of b2 – 4ac < 0, an
exception should be raised.
SECTION III: DATA STRUCTURES

Sorting and Searching
1. Implement linear search and binary search. Compare the time for searching
an element from a list of 500 random numbers.
2. Repeat the experiment for a list of 5000 integers and compare the time for
searching for an element with the two algorithms. Does increasing the
number of elements 10 times increase the running time 10 fold?
3. Implement counting sort. (Reference at the end of this appendix)
4. Implement bucket sort. (Refer to the links at the end of this appendix)
5. Implement selection sort which takes O(n log n) time.
6. Now take a list of 500 integers and compare the time for selection sort and
bucket sort.
7. Which of the two - bucket sort or counting sort - takes less time? Are they
really comparable?
8. Implement quick sort and merge sort using lists.
9. Take an array of 5000 random integers and compare the time of running of
quick sort and merge sort.
10. Can the average case complexity of quick sort be bettered?
Stacks and Queues
11. Refer to the chapter on stacks and queues and implement a dynamic stack in
which a single placeholder is added when overflow occurs.
12. Refer to the chapter on stacks and queues and implement a dynamic stack in
which the number of placeholders is doubled when overflow occurs.
13. Implement a dynamic stack in which the number of placeholders is randomly
increased when overflow occurs.
14. Using a stack convert an infix expression into a postfix expression.
15. Using stacks convert an infix expression into a prefix expression.
16. Using stacks find the nth Fibonacci term.
17. Using stacks find the number obtained by reversing the order of digits for a
given number.
18. Using queues implement priority scheduling.
19. Using queues implement First Come First Serve scheduling.
20. Using queues implement First Come First Serve with time slice.
Linked List
21. Write a program to find whether a given linked list has a cycle.
22. Write a program to join two linked lists.
23. Write a program to merge two linked lists.
24. Write a program to remove duplicate elements from a given linked list.
25. Write a program to find the second maximum element from a given linked
list.
26. Write a program to find the element greater than the mean (assume that the
linked list has only integers in the data part).
27. Write a program to find the common elements from two given linked lists.
28. Write a program to find the union of elements of two linked lists.
29. Write a program to arrange the elements of a linked list in descending order.
30. Write a program to partition a linked list as per the algorithm in the following
reference.
Graphs and Trees
A graph can be represented using a two-dimensional array. The array will

contain 0’s and 1’s. If the element at the ith row and the jth column has 1, it
indicates the presence of an edge from vertex i to j. Ask the user to enter the
number of vertices of a graph and create a two dimensional array depicting the
graph.
31. Find the number of edges in the graph. (Note that the number of 1’s in the 2-
D array is not same as the number of edges in the graph).
32. Find the vertex connected to the maximum number of edges.
33. Find if the graph has a cycle.
34. Ask the user to enter the initial vertex and the final vertex and find if there is
a path from the initial to the final vertex.
35. In the above question find whether there is more than one path from the
initial to the final vertex, in which case find the shortest path.
36. Now, in place of 1’s ask the user to enter a finite number representing the
cost of the edge from the vertex i to the vertex j. Find the shortest path from
the source vertex to all other vertices.
37. Write a program to find the spanning tree of the graph.
38. Write a program to find whether the graph is a tree.
39. A tree can be represented using a two dimensional array having n rows and
two columns. In each row the first column is i and the second column is j,
which means that there is an edge from i to j. Ask the user to enter the
requisite data and display the tree (just the list of vertices and edges
associated with them).
40. Create a binary tree using a doubly linked list. For this tree accomplish the
following tasks.
41. Write a program to implement the post order traversal of a binary tree.
42. Write a program to implement the pre order traversal of a binary tree.
43. Write a program to implement the in order traversal of the tree.
44. Check if the given tree is a binary search tree.
45. In a given binary search tree, find the leftmost node of the right sub tree of a
given node.
46. In a given binary search tree, find the rightmost node of the left sub tree of a
given node.
47. Write a program to insert an element in a binary search tree.
48. Write a program to delete a given node from a given binary search tree.
49. Write a program to create a heap from a given list.
50. Implement heap sort.
APPENDIX
D
PROBLEMS FOR PRACTICE:

SECTION 3: INTRODUCTION TO DATA STRUCTURES

Introduction
1. In linear data structures, the elements are

(a) Stored in a linear fashion
(b) Elements are accessed in sequential order
(c) Both
2. In non-linear data structures

(a) Elements are stored and accessed in a non-linear fashion
(b) Elements are accessed in a non-linear fashion
(c) Elements are stored in a non-linear fashion
3. An ADT consists of
(a) Declaration of data
(b) Declaration of operations
(c) Definition of data
(d) Definition of operations
4. The time complexity of inserting an element at a given position, in an array,
is
(a) O(n)
(b) O(n2)
(c) O(log n)
5. The time complexity of deleting an element at a given position, from an

array, requires
(a) O(n)
(b) O(n2)
(c) O(log n)
6. Which of the following is not a type of algorithmic analysis?

(a) Worst Case
(b) Best Case
(c) Average Case
(d) Boundary Case
7. The time complexity of linear search is

(a) O(n)
(b) O(n2)
(c) O(n3)

(a) (n)
(b) (n2)
(c) (n3)

(a) (n)
(b) (n2)
(c) (n3)
10. If f(n) = 3n2 + 5n + 2, Then f(n) is

(a) (n2)
(b) (n2)
(c) (n2)
11. Which one of the following is the most essential attribute of an algorithm?
(a) Correctness
(b) Finiteness
(c) Definiteness
12. Which of the following is recursive, but a non-recursive algorithm can be

created using stacks?
(a) Merge sort
(b) Quick sort
(c) Bubble sort
(d) Insertion sort
13. Which of the following is iterative?

(a) Merge sort
(b) Quick sort
(c) Bubble sort
(d) Insertion sort
14. Which of the following data structures cannot have a cycle?

(a) Directed graph
(b) Undirected graph
(c) Tree
15. On which of the following data structures does NumPy rely on?
(a) Array
(b) Trees
(c) Stacks
(d) Queues
Stacks and Queues
16. Given an expression: (a + b) – (c/d) × f. The postfix form of the expression

is ____________.
17. Given an expression: (a + b) – (c/d) × f. The prefix form of the expression is

____________.
18. Given an expression: (a/(b + c)) – d. The corresponding postfix expression

is ____________.
19. Given an expression: (a/(b + c)) – d. The prefix form of the expression is
____________.
20. Given an expression: a + ((b/c) × (d / f)). Which of the following is the

postfix form?
21. Given an expression: a + ((b/c) × (d / f)). The prefix form of the expression
is.
22. In dynamic stacks if one of the elements is added at the time of overflow,
what would the time complexity of copy operations be?
(a) O(n)
(b) O(n2)
(c) O(log n)
23. In the dynamic stacks if the number of the elements are doubled at the time
of overflow, what would the time complexity of copy operations be?
(a) O(n) time
(b) O(n2) time
(c) O(log n) time
(d) O(n log n) time
24. When reversing a string which one of the following data structures can be
used?
(a) Array
(b) Trees
(c) Stacks
(d) Queues
25. In case of linear queue what is the initial value of FRONT and REAR?
(a) FRONT = 0, REAR = 0
(b) FRONT = 0, REAR = – 1
(c) FRONT = – 1, REAR = Not Defined
(d) FRONT = –1, REAR = –1
26. Which one of the following is not possible in linear queues?

(a) FRONT = 0, REAR = 1
(b) FRONT = 3, REAR = 2
(c) FRONT = 2, REAR = 3
(d) FRONT = – 1, REAR = – 1
27. In which of the following cases value of FRONT and REAR are same?
(a) Empty queue
(b) A single element
(c) Both
28. In FIFO algorithm for scheduling in operating system which data structure is
used?
(a) Array
(b) Trees
(c) Stacks
(d) Queues
29. In spooling which one of the data structures is used?

(a) Array
(b) Trees
(c) Stacks
(d) Queues
30. Recursion requires which of the following data structures?

(a) Array
(b) Trees
(c) Stacks
(d) Queues
Linked List
31. In implementing stacks using linked lists which of the following operations
are used?
(a) Insertion at beginning
(b) Insertion at end
(c) Deletion from beginning
(d) Deletion from end
32. In implementing queues using linked lists which of the following operations
are used?
(a) Insertion at beginning
(b) Insertion at end
(c) Deletion at beginning
(d) Deletion at end
33. For a circular linked list which of the following is true?

(a) Pointer of last node points to first node
(b) There is no NULL pointer in a non-empty queue
(c) Pointer of the first node may point to the last node
While storing polynomial in a linked list a special node is created which
has two data members and a next pointer. The data members store the
coefficient and exponent of a given term.
34. Using the above representation what would be the complexity of addition of
polynomials?
(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(n log n)
35. In the above question, what would be the complexity of subtraction of

polynomials?
(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(n log n)
36. What would be the complexity of multiplication of polynomials?

(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(n log n)
37. Which of the following strategies is the most efficient for reversing a linked
list?
(a) Recursion
(b) Creating a new linked list in the order of the elements is reverse of that
of the original linked list
(c) Creating two pointers
(d) Using a temporary array
38. What is the complexity (best case) in reversing a linked list using above
strategy?
(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(n log n)
39. What is the strategy for finding a cycle in a linked list?

(a) Recursion
(b) A new linked list
(c) Using two pointers
(d) Using a temporary array
40. What would be the complexity of the above?

(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(n log n)
Trees
41. What is the height of a complete binary tree which has n nodes?
(a) n
(b) n log n
(c) log n
(d)
42. What is the number of nodes of a complete binary tree which has n levels?
(a) 2n
(b) n log n
(c) n2
(d) 2n
43. What is the complexity of searching in a complete binary tree which has n
nodes?
(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(n log n)
44. What is the worst case time complexity for searching in a skewed binary
tree?
(a) O(1)
(b) O(n)
(c) O(log n)
(d) O(n log n)
45. What is the best case complexity for searching an element in a skewed tree?
(a) O(n)
(b) O(n2)
(c) O(log n)
(d) O(1)
FIGURE # for questions 46–50
46. The maximum number of comparisons when searching an element in the

given tree (Figure #) is
(a) 4
(b) 3
(c) 5
(d) 1
47. The maximum number of comparisons when deleting an element from the
given tree (Figure _) is
(a) 1
(b) 2
(c) 3
(d) 4
48. Number of comparisons when inserting “36” in the given tree (Figure #) is
(a) 3
(b) 4
(c) 2
49. What is the index of “23” in the array representation of the given tree?
(a) 4
(b) 3
(c) 6
(d) 8
50. What is the maximum difference in the height of the left and the right
subtree in a balanced tree?
(a) 1
(b) 0
(c) 2
ANSWERS TO THE MCQ’S

1. (a)
2. (a)
3. (a, b)
4. (a)
5. (a)
6. (d)
7. (a, b, c)
8. (a)
9. (a)
10. (d)
11. (a)
12. (a, b)
13. (c, d)
14. (c)
15. (a)
22. (b)
23. (d)
24. (c)
25. (a)
26. (b)
27. (c)
28. (a)
29. (a, d)
30. (c)
31. (b, d)
32. (b, c)
33. (a)
34. (a)
35. (a)
36. (b)
37. (c)
38. (a)
39. (c)
40. (c)
41. (c)
42. (d)
43. (c)
44. (b)
45. (d)
46. (a)
47. (d)
48. (a)
49. (b)
50. (a)
E
APPENDIX
ANSWER TO THE MULTIPLE

CHOICE QUESTIONS
Chapter 1
1. (c)
2. (b)
3. (b)
4. (c)
5. (c)
6. (d)
7. (a)
8. (c)
9. (b)
10. (a)
11. (d)
12. (d)
13. (d)
14. (d)
15. (a)
Chapter 2
1. (a)
2. (b)
3. (b)
4. (d)
5. (a)
6. (b)
7. (b)
8. (b)
9. (d)
10. (b)
11. (d)
12. (d)
13. (a)
14. (a)
15. (a)
Chapter 3
1. (a)
2. (a)
3. (b)
4. (a)
5. (b)
6. (a)
7. (a)
8. (b)
9. (b)
10. (b)
Chapter 4
1. (b)
2. (d)
3. (c)
4. (a)
5. (d)
6. (d)
7. (d)
8. (a)
9. (a)
10. (b)
Chapter 5
1. (d)
2. (d)
3. (d)
4. (c)
5. (d)
6. (a)
7. (a)
8. (a)
9. (a)
10. (d)
Chapter 6
1. (a)
2. (a)
3. (a)
4. (a)
5. (b)
6. (a)
7. (d)
8. (d)
9. (d)
10. (a)
Chapter 7
1. (c)
2. (a)
3. (d)
4. (a)
5. (b)
6. (b)
7. (b)
8. (a)
9. (a)
10. (b)
11. (c)
12. (a)
13. (b)
14. (a)
15. (b)
16. (a)
17. (a)
18. (b)
19. (c)
20. (a)
21. (b)
22. (c)
23. (b)
24. (d)
25. (d)
Chapter 8
1. (a)
2. (a)
3. (c)
4. (a)
5. (b)
6. (a)
7. (b)
8. (a)
9. (b)
10. (b)
11. (b)
12. (c)
13. (a)
14. (c)
15. (d)
16. (d)
17. (d)
18. (b)
19. (b)
20. (b)
21. (a)
22. (a)
23. (a)
24. (b)
25. (b, c)
Chapter 9
1. (b, c, d)
2. (d)
3. (a)
4. (a)
5. (a)
6. (a)
7. (a)
8. (b)
9. (a)
10. (a)
11. (b)
12. (a)
13. (d)
14. (d)
15. (a)
16. (d)
17. (b)
18. (b)
19. (b)
20. (d)
Chapter 10
1. (a)
2. (d)
3. (a)
4. (c)
5. (a)
6. (c)
7. (b)
8. (a)
9. (a)
10. (a)
11. (a)
12. (a)
13. (a)
14. (c)
15. (b)
16. (a)
17. (a)
18. (a)
19. (a)
20. (b)
Chapter 11
1. (a)
2. (c)
3. (b)
4. (b)
5. (a)
6. (b)
7. (a)
8. (c)
9. (a)
10. (b)
11. (a)
12. (a)
13. (c)
14. (b)
15. (a)
Chapter 12
1. (a)
2. (a)
3. (d)
4. (b)
5. (b)
6. (a)
7. (b)
8. (b)
9. (c)
10. (b)
11. (a)
12. (a)
Chapter 13
1. (c)
2. (b)
3. (c)
4. (d)
5. (d)
6. (d)
7. (c)
8. (c)
9. (a)
10. (a)
Chapter 14
1. (d)
2. (d)
3. (a, b)
4. (c)
5. (a)
6. (b)
7. (d)
8. (d)
9. (b)
10. (a)
11. (d)
12. (d)
Chapter 15
1. (a)
2. (a)
3. (a)
4. (a)
5. (c)
6. (b)
7. (a)
8. (a)
9. (b)
10. (a)
11. (b)
12. (b)
13. (a)
14. (c)
15. (d)
Chapter 16
1. (b)
2. (c)
3. (a)
4. (c)
5. (d)
6. (b)
7. (a)
8. (a)
9. (b)
10. (b)
11. (a)
12. (b)
13. (b)
14. (a)
15. (a)
Chapter 17
1. (a, c, d)
2. (b, c, d)
3. (b)
4. (a)
5. (d)
6. (a)
7. (b)
8. (c)
9. (b)
10. (c)
Chapter 18
1. (a)
2. (a)
3. (a)
4. (a)
5. (a)
6. (c)
7. (c)
8. (c)
9. (a)
10. (a)
11. (c)
12. (d)
13. (a, b, c, d)
14. (d)
15. (b)
Chapter 19
1. (d)
2. (a)
3. (a)
4. (a)
5. (a)
6. (a)
7. (c)
8. (b)
9. (d)
10. (a)
Chapter 20
1. (b)
2. (c)
3. (a)
4. (d)
5. (a)
6. (d)
7. (d)
8. (a)
9. (a)
10. (a)
BIBLIOGRAPHY
Python
1. Mark Lutz, Learning Python, Fifth Edition, O’Reilly, 2013.

2. Stef Maruch and Aahz Maruch, Python for Dummies, John Wiley & Sons,
2006, ISBN:9780471778646.0020
3. David Beazley, Python Essential Reference, Third Edition, Sams Publishing,
USA, 2006.
4. Allen Downey, Think Python, How to Think Like a Computer Scientist,
Version 2.0.16, Green Tea Press, Needham, Massachusetts.
5. Wes McKinney, Python for Data Analysis, Wes McKinney. USA, 2013,
ISBN: 978-1-449- 31979-3.
6. Andrew Johansen, Python, The Ultimate Beginner’s Guide!
7. Wesley J. Chun, Core Python Programming, First Edition, Prentice Hall
PTR, 2000, ISBN: 0-13-026036-3, 8.
8. Peter Harrington, Machine Learning in Action, Manning Publishing
Company, 2012.
9. Richard L. Halterman, Learning to Program with Python, Copyright © 2011
Richard L. Halterman.
10. Willi Richert, Luis Pedro Coelho, Building Machine Learning Systems with
Python, Building Machine Learning Systems with Python, Packt Publishing,
2013.
Web resources
11. http://www.python.org
12. http://www.cheeseshop.python.org/
13. http://www.wiki.python.org
Data Structure and Algorithms
14. Cormen, Leiserson, Rivest, Stein, “Introduction to Algorithms,” Second

Edition, Prentice Hall of India,
15. Kleinberg, Tardos, “Algorithm Design,” Pearson, 2011.
16. Dave and Dave, “Design and Analysis of Algorithms,” Pearson, 2008.
17. Neapolitan, Naimipour, “Foundations of Algorithms,” Fourth Edition, Jones
& Barlett, 2013.
18. Horowitz et. al., “Algorithms,” Second Edition, University Press, 2007.
19. Levitin, “Introduction to Design and Analysis of Algorithms,” Perason, 2009.
20. Rajeev Motwani and P. Raghavan, “Randomized Algorithms,” Cambridge
University Press, New York (NY), 1995.
21. Williamson and Shmoys, “The Design of Approximation Algorithms,”
Cambridge University Press, 2012.
22. Christos Papadimitriou, “Computational Complexity,” 1st ed., Addison
Wesley, Chapter 11: Randomized computation, pp. 241–278, 1993.
23. Tenenbaum et. al.,” Data Structures Using C,” Pearson, 2006.
24. Horowitz, Sahini, “Fundamentals of Data Structures,” Galgotia Booksource,
1999.
25. Weiss, “Data Structure and Algorithm Analysis in C++,” Pearson, 2013.
26. Sharma, “Data Structures Using C,” Pearson, 2013.
27. Kanitkar, “Data Structures Through C,” BPB Publications,
28. Arora, Sanjeev, Barak, Boaz, “Computational complexity—A Modern
Approach,” Cambridge University Press, ISBN 978-0-521-42426-4, 2009.
29. Sipser, Michael, “Introduction to the Theory of Computation,” PWS
Publishing, ISBN 0-534- 94728-X, Section 8.2–8.3 (The Class PSPACE,
PSPACE-completeness), pp. 281–294, 1997.
30. Papadimitriou, Christos, “Computational Complexity,”1st ed., Addison
Wesley, ISBN 0-201- 53082-1. Chapter 19: Polynomial space, pp. 455–490,
1993.
31. Sipser, Michael, “Introduction to the Theory of Computation,” 2nd edition
ed., Thomson Course Technology, ISBN 0-534-95097-3. Chapter 8: Space
Complexity, 2006.
32. Jones and Pevzner, “An Introduction to Bioinformatics Algorithms,” MIT
Press.
33. Attwood, Parry Smith, Phukan, “Introduction to Bioinformatics,” Pearson,
2009.
34. Bishop, C.M., “Neural Networks for Pattern Recognition,” Oxford
University Press, Oxford, England, 1995.
35. Bishop, C.M., “Pattern Recognition and Machine Learning,” Springer-
Verlag, New York, 2008.
36. Goldberg, D.E., “Genetic Algorithms in Search, Optimization and Machine
Learning,” Addison- Wesley, Reading, MA, 1989.
37. Goldberg, D.E., “The Design of Innovation: Lessons from and for Competent
Genetic Algorithms”, Addison-Wesley, Reading, MA, 2002.
38. Stuart J. Russell and Peter Norvig, “Artificial Intelligence: A Modern
Approach,” 2 ed. Pearson Education, 2003.
39. Rich, E., Knight, K, “Artificial Intelligence,” McGraw-Hill, 1991.
40. Strang, Gilbert, “Introduction to Linear Algebra,” 4th ed. Wellesley, MA:
Wellesley-Cambridge Press, February 2009.
41. Vaˇsek Chv´atal, “Linear Programming,” W. H. Freeman & Co., 1983.
42. G. B. Dantzig, “Linear Programming and Extensions,” Princeton University
Press, 1963.
43. David Gale, “The Theory of Linear Economic Models,” McGraw-Hill, 1960.
44. Samuel Karlin, “Mathematical Methods and Theory in Games, Programming
and Economics,” Volume 1, Addison-Wesley, 1959.
45. James K. Strayer, “Linear Programming and Applications,” Springer-Verlag,
1989.
46. Brigham, E. Oran, “The Fast Fourier transform and its applications,”
Englewood Cliffs, N.J.: Prentice Hall, 1988.
47. Oppenheim, Alan V.; Schafer, R. W.; and Buck, J. R., “Discrete-time Signal
Processing. Upper Saddle River, N.J.,” Prentice Hall.
48. Smith, Steven W., “Chapter 8: The Discrete Fourier Transform,” The
Scientist and Engineer’s Guide to Digital Signal Processing,” Second ed.,
San Diego, Calif.: California Technical Publishing.
49. P. Duhamel, B. Piron, and J. M. Etcheto, “On computing the inverse DFT,”
IEEE Trans. Acoust., Speech and Sig. Processing 36 (2): 285–286, 1988.
50. Knuth, Donald, “Sorting and Searching—The Art of Computer
Programming,” Volume 3 (Second ed.). Addison–Wesley, 1998.
51. Bhasin, Harsh, Algorithms: Design and Analysis, Oxford University Press,
2015.
INDEX
A
Abstract classes, 228–229
Abstract data types (ADT), 284, 285
active_count, 442
_add_, 242, 243
add_axes, 406
add_subplot, 409
Aggregate functions, 377–381
Algorithms, 286–287
Anaconda, 7–11
Append mode, 126
Arange, 375
Arguments, 81
Arithmetic operators, 20
Arithmetic progression, 63, 110
Array, 282, 287–292
Assertion error, 266
Assignment operator, 20
Attribute error, 266
Attributes, 164–165
B
Base class, 200, 211
Behavior, 82, 164–165
Binary search tree, 354–364
_bool_, 250, 252
Bound methods, 220–221
Branches, 47
Break, 60, 61
Broadcasting, 381–385
Bubble sort, 293
C
Callable objects, 223–224
Ceil, 21
Character class, 448
Chronology, 5–6
Class, 5, 168–169, 180–181
Class diagram, 164, 165, 208
Clipping, 427–428
close(), 125, 129
cmap, 410, 424
color, 398
colorbar, 403
Command line arguments, 131–132
Complete binary tree, 350
Complex, 19
Composition, 204–208
Comprehension, 103, 115–118
Constructor, 168, 186–187
Continue, 60
Contour function, 426
Copy sign, 21
Cosine, 400, 401
Counters, 442
Cross product, 117, 118
Curdir, 129
Current_thread, 442
D
Daemon, 444
Data hiding, 170
Data structures, 281
Datatype, 373, 375
Decimal, 21
Decision making, 35, 36
Default constructor, 168, 186
Degree of a node, 351
Degree of a tree, 351
Deletion from the beginning, 329–330
Deletion of a node after a particular node, 330–331
Deletion of a node from the end, 331–336
Derived classes, 171, 200
Destructor, 168, 190–192, 253–254
Dictionary, 103, 104, 106, 107, 115
Digital image processing, 422
Django, 2
Dummy, 270
Dynamic typing, 4, 371, 372
E
Edge, 351
Efficiency, 287
Encapsulation, 169–170
EOFError, 266
Evaluation of postfix, 314
Except, 268
Exception, 262
Expect, 263
Explicit overriding, 202
F
Fabs, 22
Factorial, 22, 61
Fibonacci series, 91, 92
File access modes, 126–127
File.closed, 130
File handling mechanism, 124–125
File.mode, 130
File.name, 130
Fileno(), 129
Finally, 273–275
Findall(), 451, 452
First In First Out (FIFO), 283, 316
Floating point error, 266
Floating point numbers, 21, 374
Floor, 22
Flush(), 130
For, 144–147
Fractions, 22
Fromfile, 424
Functional, 5
Function overloading, 172, 188
Functions, 79, 165–168
G
Garbage collection, 253
Generator, 110–114
Geometric progression, 64
get construct, 46–47
get_ident(), 442
Global data, 170
Graph, 284, 349
Graphical User Interface (GUI), 6
_gt_, 249
H
Hierarchical inheritance, 213–216
Hybrid inheritance, 219
I
_idd_, 248
Identity, 19
If, 36, 37, 42
if-elif, 41
If-elif-else ladder, 42–43
if-else, 36
Image processing, 421–422
Imgshow, 403
Immutability, 26
Implicit, 231
Implicit inheritance, 231
ImportError, 266
Imread, 423
imshow(), 423
Indentation, 37, 41, 61
IndexError, 266
Indexing, 24
Infix, 313–315
Inheritance, 171, 200–204, 208–220
Inheritance tree, 226–228
__Init__, 187
In-order, 353, 359
In-order traversal, 353, 359
Insertion after a particular no, 328
Insertion at beginning, 327
Insertion at the end, 329
Instance variable, 183
Instantiation, 181
Integer, 19, 21, 163
IntendationError, 266
Interpreted language, 17
IOERROR, 126
Isalive, 443
Isatty(), 130
__Iter__, 109
Iterable object, 109–110
Iterative, 292
Iterative algorithm, 292–296
Iterators, 107–109
J
Join, 153, 443
K
KeyError, 266
Keywords, 20
Kivy, 2
L
Last in first out, 282, 283, 285, 325
_Len_, 239, 250, 252
Level of a tree, 351
Linear search, 88
Linesep, 129
Linked list, 325
Linspace, 376
List, 27
List of lists, 71–73
Literal characters, 447
Log, 401
Logical operators, 20
Logspace, 377
Looping, 59
M
Manageability, 80
Match, 448–453
Matplotlib, 395, 401, 404, 409
Matplotlib.axes, 405
Matrices, 430
Max, 428
Maximum element, 362–364
MAX_PRIORITY, 441
Mean, 428
Merge sort, 297–298
Method overriding, 179
Methods, 220–226
Min, 428
Minimum element, 362–364
Min_priority, 441
Misc.imread, 424
Misc.imsave, 424
Mixing, 3–4
Modifiers, 447
Modular Programming, 80, 162, 172
_mul_, 242–244
Multilevel inheritance, 216–219
Multiple Inheritance, 208, 219
Multiprocessing, 439
Multi-threading, 439–440
N
Ndarray, 373
Nesting, 70, 185
New, 441
_Next_, 109
No garbage collection, 253–254
None, 19
Non linear data structures, 284, 326
Normal one, 444
NotImplementedError, 267
Numarray, 372
Numbers, 20, 21
Numeric, 372
Numerical Python, 372
NumPy, 372–374
Numpy.all, 378
Numpy.any, 378
Numpy.argmax, 377
Numpy.argmin, 377
Numpy.array, 377
Numpy.max, 377
Numpy.mean, 377
Numpy.median, 377
Numpy.min, 377
Numpy.percentile, 377
Numpy.prod, 377
Numpy.std, 377
Numpy.sum, 377
Numpy.var, 377
O
Object, 169
Object oriented, 5
Object oriented paradigm, 163, 169
Objects, 30
Open, 283
Operator overloading, 237–254
Operators, 20
Object Oriented programming, 168–172, 200
Overflow, 296, 308
OverflowError, 266
Overloading, 81
Overloading binary operators, 242–246
Overriding, 201
P
Parameter, 82
Parameterized constructor, 168, 187
Pardir, 129
Parent, 351
Pathsep, 129
Patterns, 65–70
PEP (Python Enhancement Project), 6
Plot, 396
Plot_surface, 413
Polymorphism, 171–172
Portable, 4
Postfix, 313–314
Post-order, 354
Post-order traversal, 354
Power, 62, 94
Prefix, 313
Prefix to Postfix, 315
Prefix to Prefix, 315
Pre-order, 353
Pre-order traversal, 353
Primary, 282
Primary data structures, 282
Private, 170
Procedural, 4–5
Procedural language, 4–5, 161
Procedural programming, 161, 172
Protected, 170
PSF (Python Software Foundation), 6
Public, 170, 180
Pyplot, 395, 400, 403
Pyplot.figure, 409
Pyplot.GridSpec, 409
Python, 1–2
Q
QTConsole, 11
Queue, 283
Queue using linked list, 339
R
Rabbit problem, 91–94
Raise, 263, 270
Assert, 263
Random, 202
Range (m, n), 75
Range (n), 75
Rational, 21
Readline(), 128
Readlines(), 128
Read mode, 126, 130, 134
Recursion, 91–95
RecursionError, 266
Recursive algorithms, 296–298
Relational operators, 20
Reliability, 210
Re module, 447
Retstep, 376
Reusability, 172
Reversing a String, 312–313
Robustness, 2
Root, 351
Rotation, 430–432
Run, 3
Running, 441
RuntimeError, 266
S
Savefig, 396, 399
Scaling, 432–433
Scions, 2
SciPy, 2
SciPy.misc, 423
Scope, 89–91
Script mode, 18
Search, 226–228
Secondary data structures, 282
Seek(), 128
Selection sort, 294–295
Self, 187, 224
sep, 129
Sequences, 19
Sets, 19
Set_xlable, 412
Set_ylabel, 412
Set_zlabel, 412
Show, 200–202
Simple inheritance, 211–213
Sin, 400
Slicing, 25
Sort, 132
Stack, 283
Stack using a linked list, 336–339
start(), 443
Start_new_thread, 442
Std, 428
StopIteration, 266
Strings, 23–27, 143
Structured Arrays, 386–387
_sub_, 242, 243
Subplot, 407
Subplots, 405–409
Subplots_adjust, 409
Super, 224, 225
Swapping, 29
SyntaxError, 266
SystemError, 267
T
TabError, 266
tell(), 128
Ternary operator, 44–45
Threading, 441–443
Threading.local, 443
Threads, 440, 442
tofile(), 424
Translation, 429–430
Translation rotation scaling, 422
Tree, 284, 299, 349–350
Tree traversal, 353
Truediv_, 244
Try, 264, 268
Tuple, 28–30
Type, 3
Type and run, 3
Type casting, 371
Type error, 267
Types of function, 85–88
U
Unbound local error, 267
Unbound method, 220–223
Underflow, 316
UnicodeError, 267
V
Value, 19
ValueError, 267
Variables, 19
View_init, 413
Virtual, 180
W
Waiting, 441
While, 61–64
Wildcard, 454
Wireframe, 413, 414
Write(), 128
writelines(), 128, 133
Write mode, 126
X
Xlabel, 396
Xticks, 400
Y
Yield, 110, 113
Ylabel, 396
Yticks, 400
Z
Zero division error, 267

Sanet - ST 1683923537python PDF

Uploaded by

Copyright:

Available Formats

Sanet - ST 1683923537python PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sanet - ST 1683923537python PDF

Uploaded by

Copyright:

Available Formats

What does the document discuss regarding licensing and permissions?

What does the document discuss regarding licensing and permissions?

What limitations does the license impose on sharing or distributing the content?

What limitations does the license impose on sharing or distributing the content?

PYTHON

MERCURY LEARNING AND INFORMATION

Original Title and Copyright: Python for Beginners.

Publisher: David Pallai

H. Bhasin. Python Basics.

Library of Congress Control Number: 2018962670

Chapter 1: Introduction to Python

Python is a strong, procedural, object-oriented, functional language crafted in the

FIGURE 1.1 Some of the applications of Python

1.2 FEATURES OF PYTHON

1.2.2 Type and Run

1.2.5 Dynamic Typing

1.2.6 Built in Object Types

1.2.7 Numerous Libraries and Tools

In Python, the task to be accomplished becomes easy—really easy. This is

A program written in Python can run in almost every known platform, be it

In a procedural language, a program is actually a set of statements which execute

Python also supports functional programming. Moreover, Python supports

So Python is a procedural, object-oriented and functional language.

1.4 CHRONOLOGY AND USES

The continuous betterment of the language has been possible because of a

Graphical User Interface (GUI) development

1. PyDev with Eclipse

1.5 INSTALLATION OF ANACONDA

FIGURE 1.2(f) When the installation is complete, this screen appears

FIGURE 1.3 The Anaconda navigator

In Python, statements do not end with any special characters. Python

PSF: Python Software Foundation

1. Python can subclass a class made in

2. Who created Python?

3. Monty Python was

4. In Python, libraries and tools

7. There is no data type, so a code in Python is applicable to whole range of

8. Which of the following is automatic memory management?

12. What can be done using Python?

13. Python is used in

14. Python is used in

15. Which is faster?

2. Explain a few applications of Python.

3. What type of language is Python? (Procedural, object-oriented or functional)

6. Who manages Python?

7. Is Python open source or proprietary?

8. What languages can be supported by Python?

9. Explain the chronology of the development of Python.

10. Name a few editors for Python.

11. What are the features of Python?

12. What is the advantage of using Python over other languages?

13. What is Dynamic Typing?

14. Does Python have data types?

15. How is Python different from Java?

As stated earlier Python is an interpreted language, so one need not to compile

Step 1. Go to FILE NEW

Step 2. Save the file as calc.py

Step 3. Write the following code in the file

Conversely, the script can be executed by writing Python calc.py on the