Coee Break Python

50 Workouts to Kickstart Your Rapid

Code Understanding in Python

Christian Mayer

September 2018

A puzzle a day to learn, code, and play.

i
 Contents

Contents ii
1 Introduction 1
2 A Case for Puzzle-based Learning 4
2.1 Overcome the Knowledge Gap . . . . . . . 5
2.2 Embrace the Eureka Moment . . . . . . . 7
2.3 Divide and Conquer . . . . . . . . . . . . 8
2.4 Improve From Immediate Feedback . . . . 9
2.5 Measure Your Skills . . . . . . . . . . . . . 10
2.6 Individualized Learning . . . . . . . . . . . 13
2.7 Small is Beautiful . . . . . . . . . . . . . . 14
2.8 Active Beats Passive Learning . . . . . . . 16
2.9 Make Code a First-class Citizen . . . . . . 18
2.10 What You See is All There is . . . . . . . 20

ii
CONTENTS iii

3 The Elo Rating for Python 22

3.1 How to Use This Book . . . . . . . . . . . 23
3.2 The Ideal Code Puzzle . . . . . . . . . . . 25
3.3 How to Exploit the Power of Habits? . . . 26
3.4 How to Test and Train Your Skills? . . . . 27
3.5 What Can This Book Do For You? . . . . 31

4 A Quick Overview of the Python Language 36

4.1 Keywords . . . . . . . . . . . . . . . . . . 37
4.2 Basic Data Types . . . . . . . . . . . . . . 40
4.3 Complex Data Types . . . . . . . . . . . . 43
4.4 Classes . . . . . . . . . . . . . . . . . . . . 47
4.5 Functions and Tricks . . . . . . . . . . . . 50

5 Fifty Code Puzzles 54

5.1 Hello World . . . . . . . . . . . . . . . . . 55
5.2 Variables and Float Division . . . . . . . . 57
5.3 Basic Arithmetic Operations . . . . . . . . 59
5.4 Comments and Strings . . . . . . . . . . . 61
5.5 Index and Concatenate Strings . . . . . . 64
5.6 List Indexing . . . . . . . . . . . . . . . . 67
5.7 Slicing in Strings . . . . . . . . . . . . . . 69
5.8 Integer Division . . . . . . . . . . . . . . . 72
5.9 String Manipulation Operators . . . . . . 74
5.10 Implicit String Concatenation . . . . . . . 76
5.11 Sum and Range Functions . . . . . . . . . 78
5.12 Append Function for Lists . . . . . . . . . 80
5.13 Overshoot Slicing . . . . . . . . . . . . . . 82
iv CONTENTS

5.14 Modulo Operator . . . . . . . . . . . . . . 84

5.15 Branching Statements . . . . . . . . . . . 87
5.16 Negative Indices . . . . . . . . . . . . . . . 90
5.17 The For Loop . . . . . . . . . . . . . . . . 92
5.18 Functions and Naming . . . . . . . . . . . 95
5.19 Concatenating Slices . . . . . . . . . . . . 98
5.20 Arbitrary Arguments . . . . . . . . . . . . 100
5.21 Indirect Recursion . . . . . . . . . . . . . 102
5.22 String Slicing . . . . . . . . . . . . . . . . 105
5.23 Slice Assignment . . . . . . . . . . . . . . 107
5.24 Default Arguments . . . . . . . . . . . . . 109
5.25 Slicing and the len() Function . . . . . . 112
5.26 Nested Lists . . . . . . . . . . . . . . . . . 114
5.27 Clearing Sublists . . . . . . . . . . . . . . 116
5.28 The Fibonacci Series . . . . . . . . . . . . 118
5.29 The continue Statement and the Modulo
Operator . . . . . . . . . . . . . . . . . . . 121
5.30 Indexing Revisited and The Range Sequence123
5.31 Searching in Sorted Matrix . . . . . . . . . 127
5.32 Maximum Prot Algorithm . . . . . . . . 131
5.33 Bubble Sort Algorithm . . . . . . . . . . . 134
5.34 Joining Strings . . . . . . . . . . . . . . . 137
5.35 Arithmetic Calculations . . . . . . . . . . 139
5.36 Binary Search . . . . . . . . . . . . . . . . 141
5.37 Modifying Lists in Loops . . . . . . . . . . 144
5.38 The Lambda Function . . . . . . . . . . . 147
5.39 Multi-line Strings and the New-line Char-
acter . . . . . . . . . . . . . . . . . . . . . 150
CONTENTS v

5.40 Escaping . . . . . . . . . . . . . . . . . . . 152

5.41 Fibonacci Arithmetic . . . . . . . . . . . . 155
5.42 Quicksort . . . . . . . . . . . . . . . . . . 158
5.43 Unpacking Keyword Arguments with Dic-
tionaries . . . . . . . . . . . . . . . . . . . 161
5.44 Innity . . . . . . . . . . . . . . . . . . . . 164
5.45 Graph Traversal . . . . . . . . . . . . . . . 167
5.46 Lexicographical Sorting . . . . . . . . . . . 171
5.47 Chaining of Set Operations . . . . . . . . . 174
5.48 Basic Set Operations . . . . . . . . . . . . 177
5.49 Simple Unicode Encryption . . . . . . . . 180
5.50 The Guess and Check Framework . . . . . 183

6 Final Remarks 186

 1

Introduction

The great code mastersKnuth, Torvalds, and Gates

share one character trait: the ambition to learn. If you
are reading this book, you are an aspiring coder and you
seek ways to advance your coding skills. You already have
some experience in writing code, but you feel that there
is a lot to be learned before you become a master coder.
You want to read and understand code better. You want
to challenge the status quo that some of your peers un-
derstand code faster than you. Or you are already pro-
cient with another programming language like Java or
C++ but want to learn Python to become more valu-
able to the market place. Either way, you have already
proven your ambition to learn and, therefore, this book
is for you. To join the league of the great code masters,
you only have to do one thing: stay in the game.

1
2 CHAPTER 1. INTRODUCTION

The main driver for mastery is neither a character

trait, nor talent. Mastery comes from intense, struc-
tured training. The author Malcolm Gladwell formulated
the famous rule of 10,000 hours after collecting research
from various elds such as psychology and neurological
1
science. The rule states that if you have average talent,
you will reach mastery in any discipline by investing ap-
proximately 10,000 hours of intense training. Bill Gates,
the founder of Microsoft, reached mastery at a young
age as a result of coding for more than 10,000 hours. He
was committed and passionate about coding and worked
long nights to develop his skills. He was anything but an
overnight success.

There is one thing that will empower you to invest

the 10,000 hours of hard, focused work to reach mastery.
What do you think it is? As for the code masters, it's
your ambition to learn that will drive you through the
valleys of desperation on your path to mastery: complex
code, nasty bugs, and project managers pushing tight
deadlines. Nurturing your ambition to learn will pay a
rich stream of dividends to you and your family as long
as you live. It will make you a respectable member of the
society providing unique value to information technology,
automation, and digitalization. Ultimately, it will give
you strong condence. So keeping your ambition to learn
intact is the one thing you must place above all else.

1 Malcolm Gladwell Outliers: The Story of Success

 3

This book aims to be a stepping stone on your path

to becoming a Python master. It helps you to learn
faster by making use of the established principles of good
teaching. It oers you ten to twenty hours of thorough
Python training using one of the most ecient learning
techniques, called practice testing. Investing this time
will kickstart your skills to write, read, and understand
Python source code.

The idea is that you solve code puzzles that start out
simple but become more and more complex as you read
the book. In essence, you play Python interpreter and
compute the output of a code snippet in your head. Then
you check whether you were right with your guessusing
feedback and explanationsto adapt and improve your
coding skills over time. To make this idea a reality, I
developed the online coding academy Finxter.com. The
next section explains and motivates the advantages of the
Finxter method of puzzle-based learning.
 2

A Case for Puzzle-based Learning

Denition: A code puzzle is an educative

snippet of source code that teaches a single
computer science concept by activating the
learner's curiosity and involving them in the
learning process.

Before diving into practical puzzle solving, let us rst

study 10 reasons why puzzle-based learning accelerates
your learning speed and improves retention of the learned
material. There is robust evidence in psychological sci-
ence for each of these reasons. Yet, none of the existing
coding books lift code puzzles to being rst-class citi-
zens. Instead, they are mostly focused on one-directional
teaching. This book attempts to change that. In brief,
the 10 reasons for puzzle-based learning are the following.

4
2.1. OVERCOME THE KNOWLEDGE GAP 5

1. Overcome the Knowledge Gap (Section 2.1)

2. Embrace the Eureka Moment (Section 2.2)

3. Divide and Conquer (Section 2.3)

4. Improve From Immediate Feedback (Section 2.4)

5. Measure Your Skills (Section 2.5)

6. Individualized Learning (Section 2.6)

7. Small is Beautiful (Section 2.7)

8. Active Beats Passive Learning (Section 2.8)

9. Make Source Code a First-class Citizen (Section 2.9)

10. What You See is All There is (Section 2.10)

2.1 Overcome the Knowledge

Gap
The great teacher Socrates delivered complex knowledge
by asking a sequence of questions. Each question built on
answers to previous questions provided by the student.
This more than 2400 year old teaching technique is still
in widespread use today. A good teacher opens a gap be-
tween their knowledge and the learner's. The knowledge
gap makes the learner realize that they do not know the
 CHAPTER 2. A CASE FOR PUZZLE-BASED
6 LEARNING

answer to a burning question. This creates a tension in

the learner's mind. To close this gap, the learner awaits
the missing piece of knowledge from the teacher. Better
yet, the learner starts developing their own answers. The
learner craves knowledge.

Code puzzles open an immediate knowledge gap. When

looking at the code, you rst do not understand the
meaning of the puzzle. The puzzle's semantics are hid-
den. But only you can transform the unsolved puzzle into
a solved one. Look at this riddle: What pulls you down
and never lets go? Can you feel the tension? Opening
and closing a knowledge gap is a very powerful method
1
for eective learning.

Bad teachers open a knowledge gap that is too large.

The learner feels frustrated because they cannot over-
come the gap. Suppose you are standing before a river
that you must cross. But you have not learned to swim,
yet. Now, consider two rivers. The rst is the Colorado
River that carved out the Grand Canyonquite a gap.
The second is Rattlesnake Creek. The fact that you have
never heard of this river indicates that it is not too big
of an obstacle. Most likely, you will not even attempt
to swim through the big Colorado River. But you could
swim over the Rattlesnake if you stretch your abilities
just a little bit. You will focus, pep-talk yourself, and

1 The answer is Gravity.

2.2. EMBRACE THE EUREKA MOMENT 7

overcome the obstacle. As a result, your swimming skills

and your condence will grow a little bit.

Puzzles are like the Rattlesnakethey are not too

great a challenge. You must stretch yourself to solve
them, but you can do it, if you go all-out.

Constantly feeling a small but non-trivial knowledge

gap creates a healthy learning environment. Stretch your
limits, overcome the knowledge gap, and become better
one puzzle at a time.

2.2 Embrace the Eureka Moment

Humans are unique because of their ability to learn. Fast
and thorough learning has always increased our chances
of survival. Thus, evolution created a brilliant biological
reaction to reinforce learning in your body. Your brain
is wired to seek new information; it is wired to always
process data, to always learn.

Did you ever feel the sudden burst of happiness after

experiencing a eureka moment? Your brain releases en-
dorphins, the moment you close a knowledge gap. The
instant gratication from learning is highly addictive, but
this addiction makes you smarter. Solving a puzzle gives
your brain instant gratication. Easy puzzles open small,
hard puzzles, which open large knowledge gaps. Over-
come any of them and learn in the process.
 CHAPTER 2. A CASE FOR PUZZLE-BASED
8 LEARNING

2.3 Divide and Conquer

Learning to code is a complex task. You must learn a
myriad of new concepts and language features. Many
aspiring coders are overwhelmed by the complexity. They
seek a clear path to mastery.

People tend to prioritize specic activities with clearly

dened goals. If the path is not clear, we tend to drift
away toward more specic paths. Most aspiring coders
think they have a goal: becoming a better coder. Yet,
this is not a specic goal at all. So what is a specic
goal? Watching Game of Thrones after dinner, Series 2
Episode 1 is as specic as it can be. Due to the specicity,
watching Netix is more powerful than the fuzzy path of
learning to code. Hence, watching Netix wins most of
the time.

As any productivity expert will tell you: break a big

task or goal into a series of smaller steps. Finishing each
tiny step brings you one step closer to your big goal.
Divide and conquer makes you feel in control, pushing
you one step closer toward mastery. You want to become
a master coder? Break the big coding skill into a list
of sub-skillsunderstanding language features, designing
algorithms, reading codeand then tackle each sub-skill
one at a time.

Code puzzles do this for you. They break up the huge

task of learning to code into a series of smaller learning
2.4. IMPROVE FROM IMMEDIATE FEEDBACK 9

units. The student experiences laser focus on one learn-

ing task such as recursion, the for loop, or keyword ar-
guments. Don't worry if you do not understand these
concepts yetafter working through this book, you will.
A good code puzzle delivers a single idea from the au-
thor's into the student's head. You can digest one puzzle
at a time. Each puzzle is a step toward your bigger goal
of mastering computer science. Keep solving puzzles and
you keep improving your skills.

2.4 Improve From Immediate

Feedback
As a child, you learned to walk by trial and errortry,
receive feedback, adapt, and repeat. Unconsciously, you
will minimize negative and maximize positive feedback.
You avoid falling because it hurts, and you seek the ap-
proval of your parents. But not only organic life benets
from the great learning technique of trial and error. In
machine learning, algorithms learn by guessing an output
and adapting their guesses based on their correctness.
To learn anything, you need feedback such that you can
adapt your actions.

However, an excellent learning environment provides

you not only with feedback but with immediate feedback
for your actions. In contrast, poor learning environments
 CHAPTER 2. A CASE FOR PUZZLE-BASED
10 LEARNING

do not provide any feedback at all or only with a large

delay. Examples are activities with good short-term and
bad long-term eects such as smoking, alcohol, or damag-
ing the environment. People cannot control these activi-
ties because of the delayed feedback. If you were to slap
your friend each time he lights a cigarettea not overly
drastic measure to safe his lifehe would quickly stop
smoking. If you want to learn fast, make sure that your
environment provides immediate feedback. Your brain
will nd rules and patterns to maximize the reinforce-
ment from the immediate feedback.

This book oers you an environment with immediate

feedback to make learning to code easy and fast. Over
time, your brain will absorb the meaning of a code snip-
pet quicker and with higher precision this way. Learning
this skill pushes you toward the top 10% of all coders.
There are other environments with immediate feedback,
like executing code and checking correctness, but puzzle-
based learning is the most direct one: Each puzzle edu-
cates with immediate feedback.

2.5 Measure Your Skills

You need to have a denite goal to be successful. A
denite goal is a powerful motivator and pushes you to
stretch your skills constantly. The more denite and con-
crete it is, the stronger it becomes. Holding a denite
2.5. MEASURE YOUR SKILLS 11

goal in your mind is the rst and foremost step toward

its physical manifestation. Your beliefs bring your goal
into reality.

Think about an experienced Python programmer you

know, e.g., your nerdy colleague or class mate. How good
are their Python skills compared to yours? On a scale
from your grandmother to Bill Gates, where is your col-
league and where are you? These questions are dicult
to answer because there is no simple way to measure the
skill level of a programmer. This creates a severe problem
for your learning progress: the concept of being a good
programmer becomes fuzzy and diluted. What you can't
measure, you can't improve. Not being able to measure
your coding skills diverts your focus from systematic im-
provement. Your goal becomes less denite.

So what should be your denite goal when learning

a programming language? To answer this, let us travel
briey to the world of chess, which happens to provide an
excellent learning environment for aspiring players. Ev-
ery player has an Elo rating number that measures their
skill level. You get an Elo rating when playing against
other playersif you win, your Elo rating increases. Vic-
tories against stronger players lead to a higher increase
of the Elo rating. Every ambitious chess player simply
focuses on one thing: increasing their Elo rating. The
ones that manage to push their Elo rating very high,
earn grand master titles. They become respected among
 CHAPTER 2. A CASE FOR PUZZLE-BASED
12 LEARNING

chess players and in the outside world.

Every chess player dreams of being a grandmaster.

The goal is as denite as it can be: reaching an Elo of
2400 and master level (see Section 3). Thus, chess is a
great learning environmentevery player is always aware
of their skill level. A player can measure how decisions
and habits impact their Elo number. Do they improve
when sleeping enough before important games? When
training opening variants? When solving chess puzzles?
What you can measure, you can improve.

The main idea of this book, and the associated learn-

ing app Finxter.com, is to transfer this method of mea-
suring skills from the chess world to programming. Sup-
pose you want to learn Python. The Finxter website
assigns you a rating number that measures your coding
skills. Every Python puzzle has a rating number as well,
according to its diculty level. You `play' against a puz-
zle at your diculty level: The puzzle and you will have
more or less the same Elo rating so that you can enjoy
personalized learning. If you solve the puzzle, your Elo
increases and the puzzle's Elo decreases. Otherwise, your
Elo decreases and the puzzle's Elo increases. Hence, the
Elo ratings of the dicult puzzles increase over time. But
only learners with high Elo ratings will see them. This
self-organizing system ensures that you are always chal-
lenged but not overwhelmed, while you constantly receive
feedback about how good your skills are in comparison
2.6. INDIVIDUALIZED LEARNING 13

with others. You always know exactly where you stand

on your path to mastery.

2.6 Individualized Learning

The educational system today is built around the idea
of classes and courses. In these environments, all stu-
dents consume the same learning material from the same
teacher applying the same teaching methods. This tra-
ditional idea of classes and courses has a strong foun-
dation in our culture and social thinking patterns. Yet,
science proves again and again the value of individual-
ized learning. Individualized learning tailors the content,
pace, style, and technology of teaching to the student's
skills and interests. Of course, truly individualized learn-
ing has always required a lot of teachers. But paying a
high number of teachers is expensive (at least in the short
term) in a non-digital environment.

In the digital era, many fundamental limitations of

our society begin to crack. Compute servers and intel-
ligent machines can provide individualized learning with
ease. But with changing limitations, we must adapt our
thinking as well. Machines will enable truly individual-
ized learning very soon; yet society needs time to adapt
to this trend.

Puzzle-based learning is a perfect example of auto-

 CHAPTER 2. A CASE FOR PUZZLE-BASED
14 LEARNING

mated, individualized learning. The ideal puzzle stretches

the student's abilities and is neither boring nor over-
whelming. Finding the perfect learning material for each
learner is an important and challenging problem. Finx-
ter uses a simple but eective solution to solve this prob-
lem: the Elo rating system. The student solves puzzles
at their individual skill level. This book and the book's
web backend Finxter pushes teaching toward individual-
ized learning.

2.7 Small is Beautiful

The 21st century has seen a rise in microcontent. Mi-
crocontent is a short and accessible piece of valuable in-
formation such as the weather forecast, a news headline,
or a cat video. Social media giants like Facebook and
Twitter oer a stream of never-ending microcontent. Mi-
crocontent is powerful because it satises the desire for
shallow entertainment. Microcontent has many benets:
the consumer stays engaged and interested, and it is eas-
ily digestible in a short time. Each piece of microcontent
pushes your knowledge horizon a bit further. Today, mil-
lions of people are addicted to microcontent.

However, this addiction will also become a problem to

these millions. The computer science professor Cal New-
port shows in his book Deep Work that modern society
values deep work more than shallow work. Deep work is
2.7. SMALL IS BEAUTIFUL 15

a high-value activity that needs intense focus and skill.

Examples of deep work are programming, writing, or re-
searching. Contrarily, shallow work is every low-value
activity that can be done by everybody (e.g., posting
the cat videos to social media). The demand for deep
work grew with the rise of the information society; at
the same time, the supply stayed constant or decreased,
e.g., because of the addictiveness of shallow social me-
dia. People that see and understand this trend can ben-
et tremendously. In a free market, the prices of scarce
and demanded resources rise. Because of this, surgeons,
lawyers, and software developers earn $100,000 per year
and more. Their work cannot easily be replaced or out-
sourced to unskilled workers. If you are able to do deep
work, to focus your attention on a challenging problem,
society pays you generously.

What if we could marry the concepts of microcon-

tent and deep work? This is the promise of puzzle-based
learning. Finxter oers a stream of self-contained micro-
content in the form of hundreds of small code puzzles.
But instead of just being unrelated microcontent, each
puzzle is a tiny stimulus that teaches a coding concept or
language feature. Hence, each puzzle pushes your knowl-
edge in the same direction.

Puzzle-based learning breaks the bold goal, i.e., reach

the mastery level in Python, into tiny actionable steps:
solve and understand one code puzzle per day. While
 CHAPTER 2. A CASE FOR PUZZLE-BASED
16 LEARNING

solving the smaller tasks, you progress toward your larger

goal. You take one step at a time to eventually reach the
mastery level. A clear path to success.

2.8 Active Beats Passive Learning

Robust scientic evidence shows that active learning dou-
bles students' learning performance. In a study on that
matter, test scores of active learners improve by more
than one grade compared to their passive learning fellow
2
students. Not using active learning techniques wastes
your time and hinders you in reaching your full potential
in any area of life. Switching to active learning is a sim-
ple tweak that will instantly improve your performance
when learning any subject.

How does active learning work? Active learning re-

quires the student to interact with the material, rather
than simply consuming it. It is student- rather than
teacher-centric. Great active learning techniques are ask-
ing and answering questions, self-testing, teaching, and
summarizing. A popular study shows that one of the best
3
learning techniques is practice testing. In this learning
technique, you test your knowledge even if you have not

2 https://en.wikipedia.org/wiki/Active_learning#
Research_evidence
3 http://journals.sagepub.com/doi/abs/10.1177/
1529100612453266
2.8. ACTIVE BEATS PASSIVE LEARNING 17

learned everything yet. Rather than learning by doing,

it's learning by testing.

However, the study argues that students must feel

safe during these tests. Therefore, the tests must be low-
stake, i.e., students have little to lose. After the test,
students get feedback about the correctness of the tests.
The study shows that practice testing boosts long-term
retention of the material by almost a factor of 10. As it
turns out, solving a daily code puzzle is not just another
learning techniqueit is one of the best.

Although active learning is twice as eective, most

books focus on passive learning. The author delivers in-
formation; the student passively consumes the informa-
tion. Some programming books include active learning
elements by adding tests or by asking the reader to try
out the code examples. Yet, I always found this imprac-
ticable while reading on the train, on the bus, or in bed.
But if these active elements drop out, learning becomes
100% passive again.

Fixing this mismatch between research and common

practice drove me to write this book about puzzle-based
learning. In contrast to other books, this book makes
active learning a rst-class citizen. Solving code puzzles
is an inherent active learning technique. You must de-
velop the solution yourself, in every single puzzle. The
teacher is as much in the background as possiblethey
 CHAPTER 2. A CASE FOR PUZZLE-BASED
18 LEARNING

only explain the correct solution if you couldn't work it

out yourself. But before telling you the correct solution,
your knowledge gap is already ripped wide open. Thus,
you are mentally ready to digest new material.

To drive this point home, let me emphasize this ar-

gument again: puzzle-based learning is a variant of the
active learning technique named practice testing. Prac-
tice testing is scientically proven to teach you more in
less time.

2.9 Make Code a First-class

Citizen
Each grandmaster of chess has spent tens of thousands
of hours looking into a near innite number of chess posi-
tions. Over time, they develop a powerful skill: the intu-
ition of the expert. When presented with a new position,
they are able to name a small number of strong candidate
moves within seconds. They operate on a higher level
than normal people. For normal people, the position of
a single chess piece is one chunk of information. Hence
they can only memorize the position of about six chess
pieces. But chess grand masters view a whole position or
a sequence of moves as a single chunk of information. The
extensive training and experience has burned strong pat-
terns into their biological neural networks. Their brain
2.9. MAKE CODE A FIRST-CLASS CITIZEN 19

is able to hold much more informationa result of the

good learning environment they have put themselves in.

What are some principles of good learning? Let us

dive into another example of a great learning environment
this time for machines. Recently, Google's articial intel-
ligence AlphaZero has proven to be the best chess playing
entity in the world. AlphaZero uses articial neural net-
works. An articial neural network is the digital twin of
the human brain with articial neurons and synapses. It
learns by example much like a grandmaster of chess. It
presents itself a position, predicts a move, and adapts its
prediction to the extent the prediction was incorrect.

Chess and machine learning exemplify principles of

good learning that are valid in any eld you want to
master. First, transform the object to learn into a stim-
ulus that you present to yourself over and over again.
In chess, study as many chess positions as you can. In
math, make reading mathematical papers with theorems
and proofs a habit. In coding, expose yourself to lots of
code. Second, seek feedback. Immediate feedback is bet-
ter than delayed feedback. However, delayed feedback is
still much better than no feedback at all. Third, take
your time to learn and understand thoroughly. Although
it is possible to learn on-the-go, you will cut corners. The
person who prepares beforehand always has an edge. In
the world of coding, some people recommend learning by
coding practical projects and doing nothing more. Chess
 CHAPTER 2. A CASE FOR PUZZLE-BASED
20 LEARNING

grandmasters, sports stars, and intelligent machines do

not follow this advice. They learn by practicing isolated
stimuli again and again until they have mastered them.
Then they move on to more complex stimuli.

Puzzle-based learning is code-centric. You will nd

yourself staring at the code for a long time until the in-
sight strikes. This creates new synapses in your brain
that help you understand, write, and read code fast.
Placing code in the center of the whole learning pro-
cess creates an environment in which you will develop
the powerful intuition of the expert. Maximize the learn-
ing time you spend looking at code rather than at other
stimuli.

2.10 What You See is All There is

My professor of theoretical computer science used to tell
us that if we only stare long enough at a proof, the mean-
ing will transfer into our brains by osmosis. This fosters
deep thinking, a state of mind where learning is more pro-
ductive. In my experience, his staring method works
but only if the proof contains everything you need to
know to solve it. It must be self-contained.

A good code puzzle beyond the most basic level is

self-contained. You can solve it purely by staring at it
until your mind follows your eyesyour mind develops
2.10. WHAT YOU SEE IS ALL THERE IS 21

a solution based on rational thinking. There is no need

to look things up. If you are a great programmer, you
will nd the solution quickly. If not, it will take more
time but you can still nd the solutionit is just more
challenging.

My gold standard was to design each puzzle such

that it is mostly self-contained. However, to deliver on
the book's promise of training your understanding of the
Python basics, puzzles must introduce syntactical lan-
guage elements as well. But even if the syntax in a
puzzle challenges you, you should still develop your own
solutions based on your imperfect knowledge. This prob-
abilistic thinking opens the knowledge gap and prepares
your brain to receive and digest the explained solution.
After all, your goal is long-term retention of the material.
 3

The Elo Rating for Python

Pick any sport you always loved to do. How good are
you compared to others? The Elo rating answers this
question with surprising accuracy. It assigns a number
to each player that represents their skill in the sport. The
higher the Elo number, the better the player.

Let us give a small example of how the Elo rating

works in chess. Alice is a strong player with an Elo rating
of 2000 while Bob is an intermediate player with Elo 1500.
Say Alice and Bob play a chess game against each other.
Who will win the game? As Alice is the stronger player,
she should win the game. The Elo rating system rewards
players for good and punishes for bad results: the better
the result, the higher the reward. For Bob, a win, or
even a draw, would be a very good outcome of the game.
For Alice, the only satisfying result is a win. Winning

22
3.1. HOW TO USE THIS BOOK 23

against a weaker player is less rewarding than winning

against a stronger player. Thus, the Elo rating system
rewards Alice with only +3 Elo points for a win. A loss
costs her -37 Elo points, and even a draw costs her -17
points. Playing against a weaker player is risky for her
because she has much to lose but little to win.

The idea of Finxter is to view your learning as a series

of games between two players: you and the Python puz-
zle. Both players have an Elo rating. Your rating mea-
sures your current skills and the puzzle's rating reects
the diculty. On our website finxter.com, a puzzle
plays against hundreds of Finxter users. Over time, the
puzzle's Elo rating converges to its true diculty level.

Table 3.1 shows the ranks for each Elo rating level.
The table is an opportunity for you to estimate your
Python skill level. In the following, I describe how you
can use this book to test your Python skills.

3.1 How to Use This Book

This book provides a series of 50 code puzzles plus expla-
nations to test and train your Python skills. The puzzles
start from beginner level and become gradually harder
to reach intermediate level. A follow-up book covers in-
termediate to expert level. This book is perfect for users
between the beginner and the intermediate level. Yet,
24 CHAPTER 3. THE ELO RATING FOR PYTHON

Elo rating Rank

2500 World Class
2400-2500 Grandmaster
2300-2400 International Master
2200-2300 Master
2100-2200 National Master
2000-2100 Master Candidate
1900-2000 Authority
1800-1900 Professional
1700-1800 Expert
1600-1700 Experienced Intermediate
1500-1600 Intermediate
1400-1500 Experienced Learner
1300-1400 Learner
1200-1300 Scholar
1100-1200 Autodidact
1000-1100 Beginner
0-1000 Basic Knowledge

Table 3.1: Elo ratings and skill levels.

3.2. THE IDEAL CODE PUZZLE 25

even expert users can improve their speed of code un-

derstanding. No matter your current skill level, you will
benet from puzzle-based learning. It will deepen and
accelerate your understanding of basic coding patterns.

3.2 The Ideal Code Puzzle

The ideal code puzzle possesses each of the following six
properties. The puzzle

1. has a surprising result;

2. provides new information;

3. is relevant and practical;

4. delivers one main idea;

5. can be solved by thinking alone; and

6. is challenging but not overwhelming.

This was the gold standard for all the puzzles created
in this book. I did my best to adhere to this standard.
26 CHAPTER 3. THE ELO RATING FOR PYTHON

3.3 How to Exploit the Power of

Habits?
You are what you repeatedly do. Your habits determine
your success in life and in any specic area such as cod-
ing. Creating a powerful learning habit can take you
a long way on your journey to becoming a code mas-
ter. Charles Duhigg, a leading expert in the psychology
of habits, shows that each habit follows a simple process
called the habit loop. This process consists of three steps:
1
trigger, routine, and reward. First, the trigger starts the
process. A trigger can be anything such as drinking your
morning coee. Second, the routine is an action you take
when presented with the trigger. An example routine is
to solve a code puzzle. Each routine is in anticipation
of a reward. Third, the reward is anything that makes
you feel good. When you overcome a knowledge gap,
your brain releases endorphinsa powerful reward. Over
time, your habit becomes strongeryou seek the reward.

Habits with strong manifestations in these three steps

are life-changing. Invest 10% of your paycheck every
month and you will be rich one day. Get used to the
habit of solving one Python puzzle a day as you drink
your morning coeeand enjoy the endorphin dose in
your brain. Implementing this Finxter loop in your day

1 Charles Duhigg, The Power of Habit: Why We Do What We

Do in Life and Business.

3.4. HOW TO TEST AND TRAIN YOUR SKILLS?27

sets up an automatic progress toward you becoming a

better and better coder. As soon as you have established
the Finxter loop as a strong habit, it will cost you neither
a lot of time, nor energy. This is self-engineering at its
nest level.

3.4 How to Test and Train Your

Skills?
I recommend solving at least one or two code puzzles
every day, e.g., as you drink your morning coee. Then
you spend the rest of your learning time on real projects
that matter to you. The puzzles guarantee that your
skills improve over time and the real project brings you
results.

If you want to test your Python skills, use the follow-

ing simple method.

1. Track your individual Elo rating as you read the

book and solve the code puzzles. Simply write your
current Elo rating into the book. Start with an ini-
tial rating of 1000 if you are a beginner, 1500 if
you are an intermediate, and 2000 if you are an ad-
vanced Python programmer. Of course, if you al-
ready have an online rating on finxter.com, start-
ing with this rating would be the most precise op-
28 CHAPTER 3. THE ELO RATING FOR PYTHON

tion. Figure 3.4 shows ve dierent examples of

how your Elo will change while working through
the book. Two factors impact the nal rating: how
you select your initial rating and how good you per-
form (the latter being more important).

2. If your solution is correct, add the Elo points ac-

cording to the table given with the puzzle. Other-
wise, subtract the given Elo points from your cur-
rent Elo number.

Solve the puzzles in a sequential manner because they

build upon each other. Advanced readers can also solve
puzzles in the sequence they wishthe Elo rating will
work as well. The Elo rating will become more accurate
as you solve more and more puzzles. Although only an
estimate, your Elo rating is an objective measure to com-
pare your skills with the skills of others. Several Finxter
users have reported that the rating is surprisingly accu-
rate.

Use the following training plan to develop a strong

learning habit with puzzle-based learning.

1. Select a daily trigger after which you solve code

puzzles for 10 minutes. For example, decide on your
Coee Break Python, or even solve code puzzles as
you brush your teeth or sit on the train to work,
university, or school.
3.4. HOW TO TEST AND TRAIN YOUR SKILLS?29

Grand master
2500
2000
1500
Your Elo

1000
500
Beginner, 50% correct Beginner, 0% correct
0 Intermediate, 50% correct Advanced, 100% correct
Advanced, 50% correct
0 10 20 30 40 50
Number of Solved Puzzles

Figure 3.1: This plot exemplies how your Elo rating

may change while you work through the 50 code puzzles.
There are three important observations. First, no matter
how you select your initial Elo, you will converge to your
true skill level as you solve more puzzles. Second, you
will lose Elo points faster when you have a higher Elo
number. Third, your nal Elo will be anywhere between
200 and 2450 after working through this book.
30 CHAPTER 3. THE ELO RATING FOR PYTHON

2. Scan over the puzzle in a rst quick pass and ask

yourself: what is the unique idea of this puzzle?

3. Dive deeply into the code. Try to understand the

purpose of each symbol, even if it seems trivial at
rst. Avoid being shallow and lazy. Instead, solve
each puzzle thoroughly and take your time. It's
counterintuitive: To learn faster in less time, you
must stay calm and take your time and allow your-
self to dig deep. There is no shortcut.

4. Make sure you carry a pen with you and write your
solution into the book. This ensures that you stay
objectivewe all have the tendency to fake our-
selves. Active learning is a central idea of this book.

5. Look up the solution and read the explanation with

care. Do you understand every aspect of the code?
Write open questions down and look them up later,
or send them to me (info@finxter.com). I will do
everything I can to come up with a good explana-
tion.

6. Only if your solution was 100% correctincluding

whitespaces, data types, and formatting of the output
you get Elo points for this puzzle. Otherwise you
should count it as a wrong solution and swallow the
negative Elo points. The reason for this strict rule
3.5. WHAT CAN THIS BOOK DO FOR YOU? 31

is that this is the best way to train yourself to solve

the puzzles thoroughly.

As you follow this simple training plan, your skill to

see through source code quickly will improve. Over the
long haul, this will have a huge impact on your career,
income, and work satisfaction. You do not have to invest
much time because the training plan requires only 10
20 minutes per day. But you must be persistent in your
training eort. If you get o track, get right back on
track the next day. When you run out of code puzzles,
feel free to checkout Finxter.com with more than 300
hand-crafted code puzzles. I regularly publish new code
puzzles on the website as well.

3.5 What Can This Book Do For

You?
Before we dive into puzzle solving, let me anticipate and
address possible misconceptions about this book.

The puzzles are too easy/too hard. This book is for

you if you already have some experience in coding. Your
skill level in the Python programming language ranges
from beginner to intermediate. Even so, if you are al-
ready an advanced coder, this book is for you as wellif
you read it in a dierent way. Measure the time you
32 CHAPTER 3. THE ELO RATING FOR PYTHON

need to solve the puzzles and limit your solution time

to only 1020 seconds. This introduces an additional
challenge for solving the puzzles: time pressure. Solving
puzzles under time pressure sharpens your rapid code
understanding skills even more. Eventually, you will feel
that your coding intuition has improved. If the puzzles
are too hard, great. Your knowledge gap must be open
before you can eectively absorb information. Just take
your time to thoroughly understand every bit of new in-
formation. Study the cheat sheets in Chapter 4 properly.

This book is not conventionally structured by topic.

Correct, the puzzles are sorted by Elo and not struc-
tured by topic. Puzzles with a small Elo rating are easier
and more fundamental. Puzzles with a higher Elo rat-
ing are harder. To solve them, you need to combine the
fundamental learnings from the easier puzzles. Ordering
puzzles by diculty has many advantages. You can solve
puzzles in your skill level. As you are getting better, the
puzzles become harder. Finally, ordering by complex-
ity allows us to combine many topics in a single puzzle.
For example, a Python one-liner may use two topics: list
comprehension and lambda functions.

Learning to code is best done via coding on projects.

This is only part of the truth. Yes, you can improve your
skills to a certain level by diving into practical projects.
But as in every other discipline, your skills will bounce
quickly against your personal ceiling. Your ceiling is the
3.5. WHAT CAN THIS BOOK DO FOR YOU? 33

maximum skill level you are able to reach, given your

current limitations. These limitations come from a lack
of thorough understanding of basic knowledge. You can-
not understand higher-level knowledge properly without
understanding the basic building blocks. Have you ever
used machine learning techniques in your work? Without
theoretical foundations, you are doomed. Theory pushes
your ceiling upwards and gets rid of the limitations that
hold you back.

Abraham Lincoln said: Give me six hours to chop

down a tree and I will spend the rst four sharpening the
axe. Do not fool yourself into the belief that just doing
it is the most eective road to reach any goal. You must
constantly sharpen the saw to be successful in any disci-
pline. Learning to code is best done via practical coding
and investing time into your personal growth. Millions
of computer scientists enjoyed an academic education.
They know that solving hundreds or thousands of toy
examples in their studies built a strong and thorough
foundation.

How am I supposed to solve this puzzle if I do not

know the meaning of this specic Python language fea-
ture? Guess it! Python is an intuitive language. Think
about potential meanings. Solve the puzzle for each of
thema good exercise for your brain. The more you
work on the puzzle, even with imperfect knowledge, the
better you prepare your brain to absorb the puzzle's ex-
34 CHAPTER 3. THE ELO RATING FOR PYTHON

planation.

Why should I buy the book when puzzles are available

for free at Finxter. com? My goal is to remove barriers
to learning Python. Thus, all puzzles are available for
free online. This book is based on the puzzles available
at Finxter, but it extends them with more detailed and
structured information. Nevertheless, if you don't like
reading books, feel free to check out the website.

Anyway, why do some people thrive in their elds

and become valued experts while others stagnate? They
read books in their eld. They increase their value to
the marketplace by feeding themselves with valuable in-
formation. Over time, they have a huge advantage over
their peers. They get the opportunities to develop them-
selves even further. They enjoy their jobs and have much
higher work satisfaction and life quality. Belonging to the
top ten percent in your eld yields hundreds of thousands
of dollars during your career. However, there is price you
have to pay to unlock the gates to this world: you have
to invest in books and your own personal development.
The more time and money you spend on books, the more
valuable you become to the marketplace!

The Elo-based rating is not accurate. Several nxters

nd the rating helpful, fair, and accurate in comparison
to others. It provides a good indication of where one
stands in the eld of Python coders. If you feel the rating
3.5. WHAT CAN THIS BOOK DO FOR YOU? 35

is not accurate, ask yourself whether you are objective.

If you think you are, please let me know so that I have a
chance to improve this book and the Finxter back-end.
 4

A Quick Overview of the Python

Language

Before diving into the puzzles, work through the follow-

ing ve cheat sheets. They contain 80% of the Python
language features in 20% of the time. So they are de-
nitely worth your time investment.

Learn them thoroughly. Try to understand every sin-

gle line of code. And catapult your skills to the next level.
Most Python coders neglect to invest enough time into
a thorough and comprehensive understanding of the ba-
sics such as language features, data types, and language
tricks. Be dierent and absorb the examples in each of
the cheat sheets. Open up your path to become a master
coder and join the top ten percent of coders.

You can download all ve cheat sheets as concise

36
4.1. KEYWORDS 37

PDFs and post them to your wall until you know them
by heart:

https://blog.finxter.com/python-cheat-sheet/.

4.1 Keywords
All programming languages reserve certain words to have
a special meaning. These words are called keywords.
With keywords, the programmer can issue commands to
the compiler or interpreter. They let you tell the com-
puter what to do. Without keywords, the computer could
not make sense from the seemingly random text in your
code le. Note that as keywords are reserved words, you
cannot use them as variable names.

The most important Python keywords are the follow-

ing.

False True and or

not break continue class
def if elif else
for while in is
None lambda return

The next cheat sheet introduces the most important

keywords in Python. In each row, you can nd the key-
word itself, a short description, and an example of its
usage.
 CHAPTER 4. A QUICK OVERVIEW OF THE
38 PYTHON LANGUAGE

Keyword Description Code example

False​, Data values from the data False​ == (​1 ​> ​2​)
True type Boolean True​ == (​2 ​> ​1​)
and​, ​or​, Logical operators: x, y = ​True​, ​False
not (x ​and​ y)​ → both x and y (x ​or​ y) == ​True
must be True # True
(x ​or​ y)​ → either x or y (x ​and​ y) == ​False
must be True
# True
(​not​ x)​ → x must be false
(​not​ y) == ​True
# True
break Ends loop prematurely while​(​True​):
​break​ ​# no infinite loop
print(​"hello world"​)
c​ontinue Finishes current loop while​(​True​):
iteration ​continue
print(​"43"​) ​# dead code
class Defines a new class → a class​ ​Beer​:
real-world concept (object
oriented programming) ​def​ ​__init__​(self)​:
def self.content = ​1.0
Defines a new function or
class method. For latter, first
parameter ​self ​points to ​def​ ​drink​(self)​:
the class object. When self.content = ​0.0
calling class method, first
parameter is implicit. # constructor creates class
becks = Beer()

# empty beer bottle

becks.drink()
if​, Conditional program x = int(input(​"your val: "​))
elif​, execution: program starts if​ x > ​3​: print(​"Big"​)
4.1. KEYWORDS 39

else with “if” branch, tries “elif” elif​ x == ​3​: print(​"Medium"​)

branches, and finishes with else​: print(​"Small"​)
“else” branch (until one
evaluates to True).
for​, # For loop # While loop - same
while declaration semantics
for​ i ​in​ [​0​,​1​,​2​]: j = ​0
print(i) while​ j < ​3​:
print(j)
j = j + ​1
in Checks whether element is 42​ ​in​ [​2​, ​39​, ​42​] ​# True
in sequence
is Checks whether both y = x = 3
elements point to the same x​ ​is​ ​y​ ​# True
object [​3​] ​is​ [​3​] ​# False
None Empty value constant ​ ​()​:
def​ f
x = ​2
f() ​is​ ​None​ ​# True
lambda Function with no name (lambda​ x: x + ​3)(3)​ ​#
(anonymous) returns 6
return Terminates function def​ ​incrementor​(x)​:
execution and passes the ​return​ x + ​1
execution flow to the caller. incrementor(​4​) ​# returns 5
An optional value after the
return keyword specifies the
result.
 CHAPTER 4. A QUICK OVERVIEW OF THE
40 PYTHON LANGUAGE

4.2 Basic Data Types

Many programmers know basic data types as primitive
data types. They provide the primitives on which higher-
level concepts are built. A house is built from bricks.
Likewise, a complex data type is built from basic data
types. I introduce basic data types in the next cheat
sheet and complex data types in Section 4.3.

Specically, the next cheat sheet explains the three

most important (classes of ) basic data types in Python.
First, the boolean data type encodes truth values. For
example, the expression 42 > 3 evaluates to True and
1 ∈ {2, 4, 6} evaluates to False. Second, the numerical
types integer, oat, and complex numbers encode integer
values, oating point values, and complex values, respec-
tively. For example, 41 is an integer value, 41.99 is a oat
value, and 41.999+0.1∗i is a complex value (the rst part
of the equation being the real number and the second the
imaginary number). Third, the string data type encodes
textual data. An example of a string value is the Shake-
speare quote `Give every man thy ear, but few thy
voice'.
4.2. BASIC DATA TYPES 41

Data Type + Description Example

Boolean x, y = ​True​, ​False
The Boolean data type is a print(x ​and​ ​not​ y) ​# True
truth value, either ​True​ ​or print(​not​ x ​and​ y ​or​ x) ​# True
False​.
## All of those evaluate to False
These are important Boolean
operators ordered by priority if​ (​None​ ​or​ ​0​ ​or​ ​0.0​ ​or​ ​''​ ​or​ []
(from highest to lowest): ​or​ {} ​or​ set()):
not​ x​ → print(​"Dead code"​)
“if x is False, then x, else y”
## All of those evaluate to True
x ​and​ y​ → if​ (​1​ < ​2​ ​and​ ​3​ > ​2​ ​and​ ​2​ >=​2
“if x is False, then x, else y”
​and​ ​1​ == ​1​ ​and​ ​1​ != ​0​):
x ​or​ y ​ → print(​"True"​)
“if x is False, then y, else x”
Integer ## Arithmetic Operations
An integer is a positive or x, y = ​3​, ​2
negative number without print(x + y) ​# = 5
floating point (e.g. ​3​). print(x - y) ​# = 1
print(x * y) ​# = 6
Float
A float is a positive or print(x / y) ​# = 1.5
negative number with floating print(x // y) ​# = 1
point precision (e.g. print(x % y) ​# = 1s
3.14159265359​). print(-x) ​# = -3
print(abs(-x)) ​# = 3
The ‘​//​’ operator performs print(int(​3.9​)) ​# = 3
integer division. The result is
print(float(​3​)) ​# = 3.0
an integer value that is
rounded towards the smaller print(x ** y) ​# = 9
integer number (e.g. ​3​ // ​2
== ​1​).
 CHAPTER 4. A QUICK OVERVIEW OF THE
42 PYTHON LANGUAGE

String ## Indexing & Slicing

Python Strings are sequences s = ​"The youngest pope was 11 years
of characters. They are old"
immutable which means that print(s[​0​]) ​# 'T'
you can not alter the
print(s[​1​:​3​]) ​# 'he'
characters without creating a
new string. print(s[​-3​:​-1​]) ​# 'ol'
print(s[​-3​:]) ​# 'old'
The four main ways to create x = s.split() ​# string array
strings are the following. print(x[​-3​] + ​" "​ + x[​-1​] + ​" "​ +
x[​2​] + ​"s"​) ​# '11 old popes'
1. Single quotes
'Yes'
## Key String Methods
2. Double quotes
y = ​" This is lazy\t\n"
"Yes"
print(y.strip()) ​# 'This is lazy'
3. Triple quotes (multi-line)
"""Yes print(​"DrDre"​.lower()) ​# 'drdre'
We Can""" print(​"stop"​.upper()) ​# 'STOP'
4. String method s = ​"smartphone"
str(​5​) == ​'5'​ ​# True print(s.startswith(​"smart"​)) ​# True
5. Concatenation print(s.endswith(​"phone"​)) ​# True
"Ma"​ + ​"hatma"​ ​# print(​"another"​.find(​"other"​)) ​# 2
'Mahatma' print(​"cheat"​.replace(​"ch"​, ​"m"​))
# 'meat'
These are whitespace print(​','​.join([​"F"​, ​"B"​, ​"I"​]))
characters in strings. # 'F,B,I'
● Newline ​\n print(len(​"Rumpelstiltskin"​)) ​# 15
● Space ​\s print(​"ear"​ ​in​ ​"earth"​) ​# True
● Tab ​\t
4.3. COMPLEX DATA TYPES 43

4.3 Complex Data Types

In the previous section, you learned about basic data
types. These are the building blocks for complex data
types. Think of complex data types as containerseach
holding a multitude of (potentially dierent) data types.

Specically, the complex data types in this cheat sheet

are lists, sets, and dictionaries. A list is an ordered se-
quence of data values (that can be either basic or complex
data types). An example for such an ordered sequence is
the list of all US presidents:['Washington', 'Adams',
'Jefferson', ..., 'Obama', 'Trump']. In contrast,
a set is an unordered sequence of data values: { 'Trump',
'Washington', 'Jefferson', ..., 'Obama'}.
Expressing the US presidents as a set loses all order-
ing informationit's not a sequence anymore. But sets
do have an advantage over lists. Retrieving information
about particular data values in the set is much faster.
For instance, checking whether the string 'Obama' is in
the set of US presidents is blazingly fast even for large
sets. I provide the most important methods and ideas in
the following cheat sheet.
 CHAPTER 4. A QUICK OVERVIEW OF THE
44 PYTHON LANGUAGE

Complex Data Type + Example

Description
List l = [​1​, ​2​, ​2​]
A container data type print(len(l)) ​# 3
that stores a sequence of
elements. Unlike strings,
lists are mutable:
modification possible.
Adding elements [​1​, 2
​ ​, 2 ​ ​].append(​4​) ​# [1, 2, 2, 4]
to a list with append, [​1​, 2​ ​, 4 ​ ​].insert(​2​,​2​) ​# [1, 2, 2, 4]
insert, or list [​1​, 2 ​ ​, 2 ​ ​] + [​4​] ​# [1, 2, 2, 4]
concatenation. The
append operation is
fastest.
Removing elements [​1​, ​2​, ​2​, ​4​].remove(​1​) ​# [2, 2, 4]
is slower (find it first).
Reversing [​1​, ​2​, ​3​].reverse() ​# [3, 2, 1]
the order of elements.
Sorting a list [​2​, ​4​, ​2​].sort() ​# [2, 2, 4]
Slow for large lists: O(n
log n), n list elements.
Indexing [​2​, ​2​, ​4​].index(​2​)
Finds index of the first # index of element 4 is "0"
occurence of an element [​2​, ​2​, ​4​].index(​2​,​1​)
in the list. Is slow when # index of el. 2 after pos 1 is "1"
traversing the whole list.
Stack stack = [3]
Python lists can be used stack.append(​42​) ​# [3, 42]
intuitively as stack via stack.pop() ​# 42 (stack: [3])
the two list operations stack.pop() ​# 3 (stack: []​)
append() and pop().
Set basket = {​'apple'​, ​'eggs'​,
​'banana'​, ​'orange'​}
4.3. COMPLEX DATA TYPES 45

Unordered collection of same = set([​'apple'​, ​'eggs'​,

unique elements ​'banana'​, ​'orange'​])
(​at-most-once)​ . print(basket == same) ​# True
Dictionary calories = {​'apple'​ : ​52​,
A useful data structure 'banana'​ : ​89​,
for storing (key, value) 'choco'​ : ​546​}
pairs.
Reading and writing c = calories
Read and write elements print(c[​'apple'​] < c[​'choco'​]) ​# True
by specifying the key c[​'cappu'​] = ​74
within the brackets. Use
print(c[​'banana'​] < c[​'cappu'​]) ​# False
the keys() and values()
print(​'apple'​ ​in​ c.keys()) ​# True
functions to access all
keys and values of the print(​52​ ​in​ c.values()) ​# True
dictionary.
Dictionary Looping for​ k, v ​in​ calories.items():
You can access the (key, print(k) ​if​ v > ​500​ ​else​ ​None
value) pairs of a # 'chocolate'
dictionary with the
items()​ method.
Membership operator basket = {​'apple'​, ​'eggs'​,
Check with the keyword 'banana'​, ​'orange'​}
in​ ​whether the set, list, print(​'eggs'​ ​in​ basket} ​# True
or dictionary contains an
print(​'mushroom'​ ​in​ basket} ​# False
element. Set
containment is faster
than list containment.
List and Set ## List comprehension
Comprehension [(​'Hi '​ + x) ​for​ x ​in​ [​'Alice'​, ​'Bob'​,
List comprehension is 'Pete'​]]
the concise Python way # ['Hi Alice', 'Hi Bob', 'Hi Pete']
to create lists. Use
[x * y ​for​ x ​in​ range(​3​) ​for​ y ​in
brackets plus an
expression, followed by range(​3​) ​if​ x>y]
a for clause. Close with # [0, 0, 2]
 CHAPTER 4. A QUICK OVERVIEW OF THE
46 PYTHON LANGUAGE

zero or more for or if

clauses. ## Set comprehension
squares = { x**​2​ ​for​ x ​in​ [​0​,​2​,​4​] ​if​ x
Set comprehension is
< ​4​ } ​# {0, 4}
similar to list
comprehension.
4.4. CLASSES 47

4.4 Classes
Object-oriented programming is an inuential, powerful,
and expressive programming abstraction. The program-
mer thinks in terms of classes and objects. A class is a
blueprint for an object. An object contains specic data
and provides the functionality specied in the class.

Say, you are programming a game to let you build,

simulate, and grow cities. In object-oriented program-
ming, you represent all things (buildings, persons, or
cars) as objects. For example, each building object stores
data such as name, size, and price tag. Additionally,
each building provides a dened functionality such as
calculate_monthly_earnings(). This simplies read-
ing and understanding your code for other programmers.
Even more important, you can now easily divide responsi-
bilities. You code the buildings and your colleague codes
the moving cars.

In short, object-oriented programming helps you to

write readable code. By learning object orientation, your
skill of collaborating with others on complex problems
improves. The next cheat sheet introduces the most basic
concepts.
 CHAPTER 4. A QUICK OVERVIEW OF THE
48 PYTHON LANGUAGE
4.4. CLASSES 49
 CHAPTER 4. A QUICK OVERVIEW OF THE
50 PYTHON LANGUAGE

4.5 Functions and Tricks

Python is full of extra tricks and special functionality.
Learning these tricks makes you more ecient and pro-
ductive. But more importantly, these tricks make pro-
gramming easy and fun. In the next cheat sheet, I give
you the most important ones.
4.5. FUNCTIONS AND TRICKS 51

ADVANCED FUNCTIONS
map(func, iter)
Executes the function on all elements of the iterable. Example:
list(map(​lambda​ x: x[​0​], [​'red'​, ​'green'​, ​'blue'​]))
# Result: ['r', 'g', 'b']
map(func, i1, ..., ik)
Executes the function on all k elements of the k iterables. Example:
list(map(​lambda​ x, y: str(x) + ​' '​ + y + ​'s'​ , [​0​, ​2​, ​2​],
[​'apple'​, ​'orange'​, ​'banana'​]))
# Result: ['0 apples', '2 oranges', '2 bananas']
string.join(iter)
Concatenates iterable elements separated by ​string​. Example:
' marries '​.join(list([​'Alice'​, ​'Bob'​]))
# Result: 'Alice marries Bob'
filter(func, iterable)
Filters out elements in iterable for which function returns False (or 0). Example:
list(filter(​lambda​ x: ​True​ ​if​ x>​17​ ​else​ ​False​, [​1​, ​15​, ​17​,
18​])) ​# Result: [18]
string.strip()
Removes leading and trailing whitespaces of string. Example:
print(​" \n \t 42 \t "​.strip()) ​# Result: 42
sorted(iter)
Sorts iterable in ascending order. Example:
sorted([​8​, ​3​, ​2​, ​42​, ​5​]) ​# Result: [2, 3, 5, 8, 42]
sorted(iter, key=key)
Sorts according to the key function in ascending order. Example:
sorted([​8​, ​3​, ​2​, ​42​, ​5​], key=​lambda​ x: ​0​ ​if​ x==​42​ ​else​ x)
# [42, 2, 3, 5, 8]
help(func)
Returns documentation of func. Example:
 CHAPTER 4. A QUICK OVERVIEW OF THE
52 PYTHON LANGUAGE

help(str.upper()) ​# Result: '... to uppercase.'

zip(i1, i2, ...)
Groups the i-th elements of iterators i1, i2, … together. Example:
list(zip([​'Alice'​, ​'Anna'​], [​'Bob'​, ​'Jon'​, ​'Frank'​]))
# Result: [('Alice', 'Bob'), ('Anna', 'Jon')]
Unzip
Equal to: 1) unpack the zipped list, 2) zip the result. Example:
list(zip(*[(​'Alice'​, ​'Bob'​), (​'Anna'​, ​'Jon'​)]
# Result: [('Alice', 'Anna'), ('Bob', 'Jon')]
enumerate(iter)
Assigns a counter value to each element of the iterable. Example:
list(enumerate([​'Alice'​, ​'Bob'​, ​'Jon'​]))
# Result: [(0, 'Alice'), (1, 'Bob'), (2, 'Jon')]
TRICKS
python -m http.server <P>
Want to share files between your PC and your phone? Run this command in
your PC’s shell. <P> is any port number between 0–65535. Type < IP address
of PC>:<P> in the phone’s browser. Now, you can browse the files in the PC’s
directory.
Read comic
import​ antigravity
Opens the comic series xkcd in your web browser
Zen of Python
import​ this
'...Beautiful is better than ugly. Explicit is ...'
Swapping variables
This is a breeze in Python. No offense, Java! Example:
a, b = ​'Jane'​, ​'Alice'
a, b = b, a
# Result: a = 'Alice', b = 'Jane'
4.5. FUNCTIONS AND TRICKS 53

Unpacking arguments
Use a sequence as function arguments via asterisk operator *. Use a dictionary
(key, value) via double asterisk operator **. Example:
def​ ​f​(x, y, z)​:
​return​ x + y * z
f(*[​1​, ​3​, ​4​]) ​# 13
f(**{​'z'​ : ​4​, ​'x'​ : ​1​, ​'y'​ : ​3​}) ​# 13
Extended Unpacking
Use unpacking for multiple assignment feature in Python. Example:
a, *b = [​1​, ​2​, ​3​, ​4​, ​5​]
# Result: a = 1, b = [2, 3, 4, 5]
Merge two dictionaries
Use unpacking to merge two dictionaries into a single one. Example:
x={​'Alice'​ : ​18​}
y={​'Bob'​ : ​27​, ​'Ann'​ : ​22​}
z = {**x,**y}
# Result: z = {'Alice': 18, 'Bob': 27, 'Ann': 22}
 5

Fifty Code Puzzles

In the previous chapters, we have seen the benets of

puzzle-based learning. Moreover, we have revisited the
most important Python keywords, data structures, tips,
and tricks. Now take your pen, ll your cup of coee,
and let's dive into the 50 code puzzles in the book. The
puzzles are very basic in the beginning but will become
harder and harder as you proceed with the book. Again,
take your time and try to understand each and every line
until you move on to the next puzzle.

54
5.1. HELLO WORLD 55

5.1 Hello World

Puzzle 1
#############################
## id 321
## Puzzle Elo 527
## Correctly solved 86 %
#############################

print('hello world')

Puzzle 1: What is the output of this code?

Code must communicate with the outside world to

have any impact. If there is no interface, it is not worth
executing the codethe precious CPU power would be
better spent on crypto mining.

Via the print function, you connect your program to

the outside world. This function, as the name indicates,
prints a value to the standard output.

What is the standard output? You can think of it

as the environment in which your Python program lives.
Your standard output is the air around you. If you shout
Ouch! , every person in your environment can read from
your standard output that you just experienced pain.
56 CHAPTER 5. FIFTY CODE PUZZLES

The data that is printed to the standard output is

of type string. A string is a sequence of characters and
is dened via each of the three following ways: via the
single quote ('), double quote ("), or triple quote (''' and
"""). In the puzzle, we use the single quote to dene our
string 'hello world'.
Again, start with an initial rating of 1000 if you are a
beginner, 1500 if you are an intermediate, and 2000 if you
are an advanced Python programmer. If your solution
was correct, add the respective Elo dierence from the
table to your current Elo number. Otherwise, subtract it
from your current Elo number.

The correct solution 

hello world

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 41 -14
500 - 1000 16 -39
1000 - 1500 8 -47
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.2. VARIABLES AND FLOAT DIVISION 57

5.2 Variables and Float Division

Puzzle 2
#############################
## id 315
## Puzzle Elo 625
## Correctly solved 91 %
#############################

x = 55 / 11
print(x)

Puzzle 2: What is the output of this code?

The majority of people solve this puzzle correctly.

The puzzle has two goals. First, it introduces the concept
of variables. Python evaluates the result of the expres-
sion on the right side of the equation and stores it in the
variable x. After dening the variable, you can access it
at any point in the program code.

Second, it forces you to read code carefully by means

of an interesting twist: Division always returns a oating
point number. Thus, variable x stores the oat value 5.0.
The print function outputs the result as a oat and not
as an integer value 5. This is the source of most errors in
the code. People focus too much on what they mean (se-
58 CHAPTER 5. FIFTY CODE PUZZLES

mantics) and too little on how they say it (syntax). But

computers are not good yet at interpreting the meaning
of people. We have to talk to them in their language. So
you get zero points for this puzzle if your solution was
the integer value 5.

The correct solution 

5.0

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 43 -12
500 - 1000 21 -34
1000 - 1500 9 -46
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.3. BASIC ARITHMETIC OPERATIONS 59

5.3 Basic Arithmetic Operations

Puzzle 3
#############################
## id 314
## Puzzle Elo 666
## Correctly solved 75 %
#############################

x = 50 * 2 + (60 - 20) / 4
print(x)

Puzzle 3: What is the output of this code?

The Python interpreter is a powerful tool. In this

puzzle, it acts as a simple calculator. It takes a basic
mathematical expression and calculates the result.

The syntax of expressions is straightforward: use the

operators +, −, ∗ and / exactly as you have learned them
in school. The Python interpreter will handle basic rules
such as multiplication before addition for you.

Note that a common mistake here is that people write

the result as an integer instead of a oat. This can lead
to bugs in the code that are hard to nd.
60 CHAPTER 5. FIFTY CODE PUZZLES

The correct solution 

110.0

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 44 -11
500 - 1000 23 -32
1000 - 1500 9 -46
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.4. COMMENTS AND STRINGS 61

5.4 Comments and Strings

Puzzle 4
#############################
## id 313
## Puzzle Elo 691
## Correctly solved 78 %
#############################

# This is a comment
answer = 42 # the answer

# Now back to the puzzle

text = "# Is this a comment?"
print(text)

Puzzle 4: What is the output of this code?

This puzzle introduces two basic concepts. First, vari-

ables can hold strings. In fact, variables can hold any
data type. The interpreter determines the data type of
a variable at runtime. The data type of a variable can
change: you can assign a string to a variable, followed by
an integer. Second, comments in the code start with the
hash character # and end with the start of the next line.
Comments are important to improve the readability of
your code.
62 CHAPTER 5. FIFTY CODE PUZZLES

The small twist in this puzzle is the question whether

the hash character within the string literal starts a new
comment. This is not the casea comment cannot ap-
pear within a string.

There are two types of comments: block comments

and inline comments. Block comments are indented to
the same level as the commented code. Inline comments
are separated by at least two spaces on the same line as
the commented code.

The Python standard recommends to write comments

as complete sentences. Moreover, the standard discour-
ages the use of inline comments because inline comments
are often unnecessary and clutter the code.

Note that according to the standard, the second com-

ment in the puzzle is considered as bad style. Write
short and conside code and do not overuse comments.
Be aware, a friend of mine working at Google told me
that he got critized for commenting obvious statements
during the coding interview.

The correct solution 

# Is this a comment?
5.4. COMMENTS AND STRINGS 63

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 45 -10
500 - 1000 24 -31
1000 - 1500 9 -46
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

64 CHAPTER 5. FIFTY CODE PUZZLES

5.5 Index and Concatenate

Strings

Puzzle 5
#############################
## id 331
## Puzzle Elo 742
## Correctly solved 63 %
#############################

x = 'silent'
print(x[2] + x[1] + x[0]
+ x[5] + x[3] + x[4])

Puzzle 5: What is the output of this code?

This puzzle introduces a powerful tool for your Python

toolbox: indexing. Make sure you feel comfortable using
it because many advanced puzzles build on your proper
understanding of indexing.

In Python, you can access every character in the string

by using an integer value that denes the position of the
character in the string. We call this integer value an
index.
5.5. INDEX AND CONCATENATE STRINGS 65

If the string has ve characters as in the example, the

indices of these characters are as follows.

String s: s i l e n t
Index: 0 1 2 3 4 5

You can index any character using the square bracket

notation [] with their respective position values. Many
programming novices are confused by the fact that the
rst element in a sequence has index 0. Therefore, you
access the rst character 's' with the expression s[0]
and the third character 'l' with the expression s[2].
The plus operator + is context sensitive. It calculates
the mathematical sum for two given numerical values but
appends strings for two given string values. For example,
the expression 'a' + 'b' returns a new string 'ab'.
With this information, you are now able to determine
how string s is reordered using indexing notation and the
`+' operator for strings.

A small note in case you were confused. There is no

separate character type in Python; a character is a string
of size one.

The correct solution 

listen
66 CHAPTER 5. FIFTY CODE PUZZLES

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 45 -10
500 - 1000 27 -28
1000 - 1500 10 -45
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.6. LIST INDEXING 67

5.6 List Indexing

Puzzle 6
#############################
## id 337
## Puzzle Elo 745
## Correctly solved 91 %
#############################

squares = [1, 4, 9, 16, 25]

print(squares[0])

Puzzle 6: What is the output of this code?

This puzzle introduces the simple but powerful list

data structure in Python. You have to search very hard
to nd an algorithm that doesn't use a list. Many famous
algorithms such as quicksort are based only on a single
list as their core data structure.

Wikipedia denes a list as an abstract data type that

1
represents a countable number of ordered values. The
data type is abstract because you can use lists indepen-
dently of the concrete data type(s) of the list elements.

1 https://en.wikipedia.org/wiki/List_(abstract_data_
type)
68 CHAPTER 5. FIFTY CODE PUZZLES

The Python way of handling lists and list access is

simple and clean. Create a list by writing comma-separated
values between the opening and closing square brackets.

In the Java programming language, you must use re-

dundant natural language function calls such as get(i)
to access a list value. In Python, this is much easier.
You access the i-th element in a list lst with the intu-
itive bracket notation lst[i]. This notation is consistent
for all compound data types such as strings and arrays.

This leads to small and repeated time savings during

programming. The time savings of millions of developers
add up to a strong collective argument for Python.

The correct solution 

1

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 45 -10
500 - 1000 27 -28
1000 - 1500 10 -45
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.7. SLICING IN STRINGS 69

5.7 Slicing in Strings

Puzzle 7
#############################
## id 336
## Puzzle Elo 778
## Correctly solved 72 %
#############################

word = "galaxy"
print(len(word[1:]))

Puzzle 7: What is the output of this code?

More than one out of four Finxter users cannot solve

this puzzle. There are two concepts that are novel for
them: the len() function and slicing.

The len() function is a handy tool to get the length

of built-in Python data types such as strings, lists, dic-
tionaries, or tuples. Learn it now and make your future
life easier.

Slicing is a Python-specic concept for accessing a

range of values in sequence types such as lists or strings.
It is one of the most popular Python features. Under-
standing slicing is one of the key requirements for un-
70 CHAPTER 5. FIFTY CODE PUZZLES

derstanding most existing Python code bases. In other

words, the time you invest now in mastering slicing will
be repaid a hundredfold during your career.

The idea of slicing is simple. Use the bracket nota-

tion to access a sequence of elements instead of only a
single element. You do this via the colon notation of
[start:end]. This notation denes the start index (in-
cluded) and the end index (excluded). Note that forget-
ting that the end index is always excluded in sequence
operators is a very common source of bugs.

For the sake of completeness, let me quickly explain

the advanced slicing notation [start:end:step]. The
only dierence to the previous notation is that it allows
you to specify the step size as well. For example, the
command 'python'[:5:2] returns every second charac-
'pto'.
ter up to the fourth character, i.e., the string

The correct solution 

5
5.7. SLICING IN STRINGS 71

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 29 -26
1000 - 1500 10 -45
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

72 CHAPTER 5. FIFTY CODE PUZZLES

5.8 Integer Division

Puzzle 8
#############################
## id 316
## Puzzle Elo 781
## Correctly solved 69 %
#############################

x = 50 // 11
print(x)

Puzzle 8: What is the output of this code?

When I started to learn Python 3, I used to be con-

fused about the semantics of dividing two integers. Is
the result a oat or an integer value? The reason for my
confusion was a nasty Java bug that I once found in my
code. The code was supposed to perform a simple divi-
sion of two integers and return a oat parameter value
between zero and one. But Java uses integer division,
i.e., it skips the remainder. Thus, the value was always
either zero or one, but took never a value in-between. It
took me days to gure that out.

Save yourself the debugging time by memorizing the

following rule once and for all. The // operator per-
5.8. INTEGER DIVISION 73

forms integer (oor) division and the / operator performs

oat (true) division. An example for oor division is 50
// 11 = 4. An example for true division is 50 / 11 =
4.545454545454546.
Note that oor division always rounds down, i.e., 3
// 2 == 1 and -3 // 2 == -2.
Although the puzzle seems simple, more than twenty
percent of the Finxter users cannot solve it.

The correct solution 

4

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 29 -26
1000 - 1500 10 -45
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

74 CHAPTER 5. FIFTY CODE PUZZLES

5.9 String Manipulation

Operators

Puzzle 9
#############################
## id 327
## Puzzle Elo 786
## Correctly solved 60 %
#############################

print(3 * 'un' + 'ium')

Puzzle 9: What is the output of this code?

Python has powerful built-in capabilities for string

manipulation. Web companies like Google love Python
because it is a perfect t for the text-based World Wide
Web. The puzzle explains two basic string manipulation
operators. The + operator concatenates two strings. The
* operator concatenates a string to itself repeatedly. The
standard arithmetic rules apply to these operators: mul-
tiplication rst, then addition.
5.9. STRING MANIPULATION OPERATORS 75

The correct solution 

unununium

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 30 -25
1000 - 1500 10 -45
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

76 CHAPTER 5. FIFTY CODE PUZZLES

5.10 Implicit String

Concatenation
Puzzle 10
#############################
## id 328
## Puzzle Elo 794
## Correctly solved 74 %
#############################

x = 'py' 'thon'
print(x)

Puzzle 10: What is the output of this code?

A well-designed puzzle conveys one single point that

surprises the reader. This puzzle introduces a language
feature that surprised me when I rst saw it. The Python
interpreter automatically concatenates two strings that
are next to each other. Think about all the pluses + you
could save!

Just kidding: forget this trick immediately. Code is

read much more often than it is written and these tricks
will confuse some readers of your code. My editor even
recommended skipping this puzzle because it may be con-
5.10. IMPLICIT STRING CONCATENATION 77

fusing. Although he is totally right, I still keep it in the

book because I think you may nd it interestingand
because you may also want to understand other people's
dirty code.

The correct solution 

python

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 30 -25
1000 - 1500 10 -45
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

78 CHAPTER 5. FIFTY CODE PUZZLES

5.11 Sum and Range Functions

Puzzle 11
#############################
## id 93
## Puzzle Elo 815
## Correctly solved 76 %
#############################

print(sum(range(0, 7)))

Puzzle 11: What is the output of this code?

Do you know the story of the brilliant mathematician

Carl Friedrich Gauss? When 8-year old Gauss went to
school, his math teacher sought a break. He told his class
to solve the problem of adding all consecutive numbers
from 1 to 100: 1 + 2 + 3 + ... + 100. But as little Gauss
promptly reported the solution, the break was over before
it began. Surprised (and a bit grumpy as the story goes),
the teacher asked the boy how he had come up with a
solution so quickly. Gauss explained his simple solution.
He organized the sequence into pairs of numbers each
summing up to 101: 1 + 100, 2 + 99, 3 + 98, ..., 50 + 51.
There are 50 such pairs, so the total result was 50 ∗ 101 =
5050.
5.11. SUM AND RANGE FUNCTIONS 79

Yet, the modern-time little Gauss would be even lazier.

He would type the following one-liner into his mobile
Python app: sum(range(1,101)). The range function
returns a sequence starting from the rst value (inclu-
sive) and ending at the second value (exclusive). The
sum function sums up the values of this sequence. Com-
bining both functions sums up the sequence from 1100.
Although your computer uses a brute-force approach, it
computes the result faster than any humandumb, but
blazingly fast!

The correct solution 

21

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 31 -24
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

80 CHAPTER 5. FIFTY CODE PUZZLES

5.12 Append Function for Lists

Puzzle 12
#############################
## id 341
## Puzzle Elo 821
## Correctly solved 71 %
#############################

cubes = [1, 8, 27]

cubes.append(4 ** 3)
print(cubes)

Puzzle 12: What is the output of this code?

This puzzle shows how you can add a new value to the
end of the list using the append() function. Before ap-
pending, the Python interpreter evaluates the expression
given within the brackets. Recall that the ** operator
returns the power function, i.e., 4 ** 3 reads four to the
power of three.

The correct solution 

[1, 8, 27, 64]
5.12. APPEND FUNCTION FOR LISTS 81

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 32 -23
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

82 CHAPTER 5. FIFTY CODE PUZZLES

5.13 Overshoot Slicing

Puzzle 13
#############################
## id 335
## Puzzle Elo 829
## Correctly solved 83 %
#############################

word = "galaxy"
print(word[4:50])

Puzzle 13: What is the output of this code?

This puzzle introduces a special feature of slicing. As

a recap, Python slicing means to access a subsequence
of a sequence type using the notation [start:end]. We
show here that slicing is robust even if the end index over-
shoots the maximal sequence index. So the big take away
from this puzzle isthat nothing unexpected happens if
slicing overshoots sequence indices.

The correct solution 

xy
5.13. OVERSHOOT SLICING 83

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 32 -23
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

84 CHAPTER 5. FIFTY CODE PUZZLES

5.14 Modulo Operator

Puzzle 14
#############################
## id 317
## Puzzle Elo 835
## Correctly solved 62 %
#############################

x = 51 % 3
print(x)

Puzzle 14: What is the output of this code?

When I studied computer science, the professors pushed

us to learn the theory behind modulo operations and
residual classes. But many of us lacked motivation to do
so. We could not see why calculating the remainder of
the division, i.e., modulo, is such an important concept.

Yet, many practical code projects later, I must ad-

mit that modulo plays a role in many of them. Learning
modulo is not optional. Suppose your code has a main
loop and you want to execute a monitoring function each
thousandth iteration i. Modulo is your friend here: sim-
ply use the fact that for every thousandth iteration, the
result of i%1000 is zero.
5.14. MODULO OPERATOR 85

Learning these small code patterns is the key to be-

coming a great coder. You must know them by heart,
without much thinking. This frees your mental energy
and allows you to focus on the big picture. You will pro-
duce better and more meaningful code. In fact, one of
the main ideas of the website Finxter.com is to burn
these small code patterns into your head.

The correct solution 

0

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 32 -23
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

86 CHAPTER 5. FIFTY CODE PUZZLES
5.15. BRANCHING STATEMENTS 87

5.15 Branching Statements

Puzzle 15
#############################
## id 148
## Puzzle Elo 845
## Correctly solved 36 %
#############################

def if_confusion(x, y):

if x > y:
if x - 5 > 0:
x = y
if y == y + y:
return "A"
return "B"
elif x + y > 0:
while x > y: x -= 1
while y > x: y -= 1
if x == y:
return "E"
else:
x *= 2
if x == y:
return"F"
return "G"
else:
if x - 2 > y - 4:
x_old = x
x = y * y
y = 2 * x_old
if (x - 4) ** 2 > (y - 7) ** 2:
return "C"
return "D"
return "H"

print(if_confusion(3, 7))

Puzzle 15: What is the output of this code?

88 CHAPTER 5. FIFTY CODE PUZZLES

Now it is getting interesting! When I made this puz-

zle, I thought that it might be too simple. But look at
the numbers: only 36% of our users solved it. An inter-
esting observation is that the puzzle still has a low Elo
rating. This indicates that nxters with higher Elo can
solve it easily. Hence, these intermediate to advanced
coders have low error rates and push it down the Elo
ladder. But nxters with lower Elo ratings struggle with
the puzzle.

Here are a few tips for the latter group. Never let
the sheer mass of code intimidate you. You do not have
to read each and every line to adhere to any kind of
perfectionism. Your computer does not execute the code
strictly from top to bottom and you shouldn't as well.
Instead, start where the programm execution starts: at
the bottom with the function call if_confusion(3, 7).
Now you know that x=3 and y=7. Then you proceed to
do what the interpreter does. As x>y is false, you can
skip the whole upper part of the function. Similarly, you
can skip the if branch for x-2>y-4. It's easy to see that
the function returns 'H'.

The correct solution 

H
5.15. BRANCHING STATEMENTS 89

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 33 -22
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

90 CHAPTER 5. FIFTY CODE PUZZLES

5.16 Negative Indices

Puzzle 16
#############################
## id 332
## Puzzle Elo 848
## Correctly solved 54 %
#############################

x = 'cool'
print(x[-1] + x[-2]
+ x[-4] + x[-3])

Puzzle 16: What is the output of this code?

You can index single characters in strings using the

bracket notation. The rst character has index 0, the
second index 1, and so on. Did you ever want to access
the last element in a string? Counting the indices can
be a real pain for long strings with more than 810 char-
acters. But no worries, Python has a language feature
for this. Instead of starting counting from the left, you
can also start from the right. Access the last character
with the negative index -1, the second last with the index
-2, and so on. In summary, there are two ways to index
sequence positions, from the left and from the right.
5.16. NEGATIVE INDICES 91

The correct solution 

loco

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 33 -22
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

92 CHAPTER 5. FIFTY CODE PUZZLES

5.17 The For Loop

Puzzle 17
#############################
## id 348
## Puzzle Elo 858
## Correctly solved 67 %
#############################

words = ['cat', 'mouse']

for word in words:
print(len(word))

Puzzle 17: What is the output of this code?

Repetition is everywhere. The sun goes up every

morning and after winter comes spring. As coders, we
model and simulate the real world and create our own
worlds with our own laws and repetitions. Suppose you
want to program a web server that repeats forever the
following behavior. Wait for a user request and answer
it. How can you program the web server to repeat this
behavior thousands of times?

The naive approach is to put the sequence of steps

into the source code itself. In other words, copy and paste
the sequence of steps thousands of times. Yet, repeated
5.17. THE FOR LOOP 93

code is redundant and hard to read, debug, and maintain.

As programmers, we should avoid redundant code at all
costs.

The Python for loop statement is a way out of re-

dundant code. You write code only once and put it into
dierent contexts. For example, the loop variable (i.e.,
word in the puzzle) accounts for the dierent contexts of
loop executions. In the puzzle, the variable word takes
rst the value `cat' and second the value `mouse'.

Among the ingredients that make a programming lan-

guage powerful are control ow statements. The Python
for loop is one such control ow statement. It repeats
execution of the code body for all sequence elements
iterating over all elements in the order of the sequence.
In the puzzle, we print out the length of each word.

The correct solution 

3
5
94 CHAPTER 5. FIFTY CODE PUZZLES

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 46 -9
500 - 1000 34 -21
1000 - 1500 11 -44
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.18. FUNCTIONS AND NAMING 95

5.18 Functions and Naming

Puzzle 18
#############################
## id 358
## Puzzle Elo 899
## Correctly solved 61 %
#############################

def func(x):
return x + 1

f = func
print(f(2) + func(2))

Puzzle 18: What is the output of this code?

Too much redundant code indicates poor program-

ming style. So how to avoid redundant code? Use func-
tions. Functions make code more general. Suppose you
want to calculate the square root of 145. You could either
calculate it for the specic value 145 or dene a function
that calculates the square root for any value x.
We say that a function encapsulates a sequence of
program instructions. The ideal function solves a single
semantic high-level goal. For instance, you can encapsu-
96 CHAPTER 5. FIFTY CODE PUZZLES

late a complex task into a function, such as searching the

web for specic keywords. In this way, the complex task
becomes a simple one-liner: calling the function. Func-
tions enable others to reuse your code and allow you to
reuse other people's code. You are standing on the shoul-
ders of giants.

You can dene a function with the keyword def, fol-

lowed by a name and the arguments of the function.
The Python interpreter maintains a symbol table that
stores all function denitions, i.e., mappings from func-
tion names to function objects. In this way, the inter-
preter can relate each occurrence of the function name
to the dened function object. Just remember: a single
function object can have zero, one, or even many names.

In the puzzle, we assign the function object to the

name func and then reassign it to the new name f. We
then use both the names in the code. Upon the function
call, the Python interpreter will nd the function in the
symbol table and execute it. This can make your code
more readable when calling the same function in dierent
contexts.

The correct solution 

6
5.18. FUNCTIONS AND NAMING 97

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 36 -19
1000 - 1500 12 -43
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

98 CHAPTER 5. FIFTY CODE PUZZLES

5.19 Concatenating Slices

Puzzle 19
#############################
## id 334
## Puzzle Elo 954
## Correctly solved 45 %
#############################

word = "galaxy"
print(word[:-2] + word[-2:])

Puzzle 19: What is the output of this code?

This puzzle revisits the important concept of slicing.

Slicing is one of the most popular features in Python.
Understand the term and concept of slicing and you are
at least among the intermediate Python programmers.

Slicing, like indexing, retrieves specic characters from

a sequence such as a string. But while indexing retrieves
only a single character, slicing retrieves a whole substring
within an index range.

Use the bracket notation for slicing with the start and
end position identiers. For example, word[i:j] returns
the substring starting from index i (included) and ending
5.19. CONCATENATING SLICES 99

in index j (excluded).

You can also skip the position identier before or af-

ter the slicing colon. This indicates that the slice starts
from the rst or last position, respectively. For example,
word[:i] + word[i:] returns the same string as word.

The correct solution 

galaxy

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 38 -17
1000 - 1500 14 -41
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

100 CHAPTER 5. FIFTY CODE PUZZLES

5.20 Arbitrary Arguments

Puzzle 20
#############################
## id 365
## Puzzle Elo 1005
## Correctly solved 57 %
#############################

def func(a, *args):

print(a)
for arg in args:
print(arg)

func("A", "B", "C")

Puzzle 20: What is the output of this code?

Suppose you want to create a function that allows an

arbitrary number of arguments. An example is recog-
nizing faces in images where each image consists of one
or more pixel arrays. You achieve this by prexing the
function argument with the asterisk operator (or star op-
erator), e.g., *pixels. Now, you can pass a tuple or a
list as a function argument, which you can access via
indexing or iteration in a loop.

You can combine both types of parameters in a func-

5.20. ARBITRARY ARGUMENTS 101

tion: normal positional parameters (e.g., a in the puzzle)

and an arbitrary length parameter list (e.g., *args in the
puzzle). If you call the function with many arguments,
the interpreter lls in name slot(s) for normal positional
arguments rst. The arbitrary argument list handles the
rest of the arguments.

The correct solution 

A
B
C

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 40 -15
1000 - 1500 15 -40
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

102 CHAPTER 5. FIFTY CODE PUZZLES

5.21 Indirect Recursion

Puzzle 21
#############################
## id 76
## Puzzle Elo 1032
## Correctly solved 54 %
#############################

def ping(i):
if i > 0:
return pong(i - 1)
return "0"

def pong(i):
if i > 0:
return ping(i - 1)
return "1"

print(ping(29))

Puzzle 21: What is the output of this code?

Recursion is a powerful tool in your coding toolbox.

Understanding it is a key skill on your path to mastery.
So what is recursion? Stephen Hawking used a concise
explanation: to understand recursion, one must rst un-
derstand recursion.
5.21. INDIRECT RECURSION 103

This puzzle uses indirect recursion: function f calls

function g which calls function f. Each function call
solves a slightly easier problem. In recursive problem
solving, a function knows the result for some base cases
(i.e., the naive solutions). It breaks a complex problem
into a combination of less complex subproblems. As the
subproblems are getting easier, they nally reach the base
cases. These are the least complex subproblems and we
know their solutions. The idea is to build the solution of
the complex problem from the solutions of the subprob-
lems.

So when you call ping(29), the ping function reduces

this question topong(28)an easier problem. The call-
ing function ping waits for pong to return a solution.
But pong asks back ping(27) and waits for a solution.
On a higher level, ping receives odd and pong even ar-
gument values for the initial input i=29. Thus, the last
call is pong(0), which returns 1. Each calling function
is waiting for the result of the called function. Each call-
ing function receives the value 1 and returns it to its
parent calling function. Finally, the top-most functional
instance ping(29) returns the value 1 as the nal result.

The correct solution 

1
104 CHAPTER 5. FIFTY CODE PUZZLES

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 41 -14
1000 - 1500 16 -39
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.22. STRING SLICING 105

5.22 String Slicing

Puzzle 22
#############################
## id 333
## Puzzle Elo 1038
## Correctly solved 53 %
#############################

word = "bender"
print(word[1:4])

Puzzle 22: What is the output of this code?

The language feature slicing does not only apply to

lists, but also to strings. As both lists and strings are se-
quencing types, slicing is only one among several similari-
ties. For example, you can also iterate over the characters
in a string using the for loop (e.g., for c in word).
Only half of the Finxter users can solve this puzzle.
The main problem is to identify the correct end index of
the slice. Recap: the end index is not included in the
slice. Here is how you can nd the correct solution (in
bold).
106 CHAPTER 5. FIFTY CODE PUZZLES

b e n d e r
0 1 2 3 4 5

The correct solution 

end

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 41 -14
1000 - 1500 17 -38
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.23. SLICE ASSIGNMENT 107

5.23 Slice Assignment

Puzzle 23
#############################
## id 342
## Puzzle Elo 1104
## Correctly solved 52 %
#############################

customers = ['Marie', 'Anne', 'Donald']

customers[2:4] = ['Barack', 'Olivia', 'Sophia']
print(customers)

Puzzle 23: What is the output of this code?

A great coder seeks the cleanest and shortest way

to accomplish their goals. This puzzle demonstrates a
Python trick that I found very useful: slice assignments.

Suppose you work in a biotech startup on DNA se-

quence modeling. You maintain dierent DNA sequences
as lists of string values. To simulate recombinations of
DNA sequences, you change subsequences of the list on a
regular basis. In this case, slicing is your best friend: It
helps you to read specic subsequences. Moreover, slice
assignments enable you to replace, append, or clear whole
subsequences.
108 CHAPTER 5. FIFTY CODE PUZZLES

In the puzzle, we have a list of customers that are

partially replaced by new customers. The puzzle shows
how the length of the original sequence may change due
to the slice assignment. The slice assignment inserts a
list of three customers into the customer list. A beautiful
way to clear the list is: customers[:] = [].

The correct solution 

[`Marie', `Anne', `Barack', `Olivia',
`Sophia']

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 43 -12
1000 - 1500 20 -35
1500 - 2000 8 -47
>2000 8 -47

Your current Elo rating 

5.24. DEFAULT ARGUMENTS 109

5.24 Default Arguments

Puzzle 24
#############################
## id 360
## Puzzle Elo 1152
## Correctly solved 50 %
#############################

def ask(prompt, retries=4, output='Error'):

for _ in range(retries):
response = input(prompt).lower()
if response in ['y', 'yes']:
return True
if response in ['n', 'no']:
return False
print(output)

print(ask('Want to know the answer?', 5))

Puzzle 24: Is ask('Want to know the answer?', 5) a

valid function call?

This puzzle introduces the concept of default argu-

ments in Python.

Suppose you have created a Python command line

110 CHAPTER 5. FIFTY CODE PUZZLES

tool for your business. The tool requires user conrma-

tion for dierent activities like writing or deleting les.

To avoid redundant code, you have implemented a

generic function that handles the interaction with the
user. The default behavior should consist of three steps:
(1) You ask (prompt) the user a yes/no question; (2) the
user enters some response; (3) as long as the response
is invalid, the function repeats up to four timeseach
time printing an error message `Error'. The number of
repetitions and the reminder should be customizable via
the parameters.

To achieve this, you can specify default arguments as

given in the puzzle. You can use the default parameters
by calling ask('Hi?'). Or you can overwrite them in the
order of their denition (one, several, or all parameters).

Also did you notice that single underscore is a valid

name in Python? By convention, you can use it as a
throw-away namewhen you don't really need to access
the actual value. In the puzzle, we ask the user four times
but do not need to know how often we have already asked.

It is interesting that only 50% of all Finxter users

solve this puzzle correctly. That's no better than random
guessing. Partial replacement of default arguments is a
new feature to most users. Is it new to you? You have
to master these basic language features before you can
climb to the level of an advanced coder.
5.24. DEFAULT ARGUMENTS 111

The correct solution 

Yes

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 44 -11
1000 - 1500 22 -33
1500 - 2000 9 -46
>2000 8 -47

Your current Elo rating 

112 CHAPTER 5. FIFTY CODE PUZZLES

5.25 Slicing and the len()

Function
Puzzle 25
#############################
## id 344
## Puzzle Elo 1211
## Correctly solved 44 %
#############################

letters = ['a', 'b', 'c', 'd']

print(len(letters[1:-1]))

Puzzle 25: What is the output of this code?

The goal of this puzzle is to deepen your understand-

ing of the important concept of slicing.

Yet, it turned out to be more a test of thoroughness

than anything else. The majority of users cannot solve
this puzzleone of the most common errors is to overlook
the word len().
The built-in function len() returns the length of a
sequence object such as a string or a list. In the puzzle,
we return the length of the list after cutting the head and
the tail. An illuminating example for lack of thorough-
5.25. SLICING AND THE LEN() FUNCTION 113

ness, which is also the major source of bugs in your code.

Ask any professional programmer!

The correct solution 

2

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 45 -10
1000 - 1500 25 -30
1500 - 2000 9 -46
>2000 8 -47

Your current Elo rating 

114 CHAPTER 5. FIFTY CODE PUZZLES

5.26 Nested Lists

Puzzle 26
#############################
## id 345
## Puzzle Elo 1238
## Correctly solved 39 %
#############################

a = ['a', 'b']
n = [1, 2]
x = [a, n]
print(x[1])

Puzzle 26: What is the output of this code?

Many practical code projects use lists containing not

only primitive but also complex data types. Examples
of primitive data types are integers, strings, or oats.
Examples of complex data types are customer objects,
camera events, or even lists.

In the puzzle, we show the latter: there is a nested

list that contains two other lists. Accessing an element
of this list using the index notation returns a list itself.

Python is a dynamically typed programming language.

Hence, there can be hybrid lists containing dierent data
5.26. NESTED LISTS 115

types. However, this is not too common.

The correct solution 

[1, 2]

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 45 -10
1000 - 1500 27 -28
1500 - 2000 9 -46
>2000 8 -47

Your current Elo rating 

116 CHAPTER 5. FIFTY CODE PUZZLES

5.27 Clearing Sublists

Puzzle 27
#############################
## id 343
## Puzzle Elo 1248
## Correctly solved 47 %
#############################

letters = ['a', 'b', 'c',

'd', 'e', 'f', 'g']
letters[1:] = []
print(letters)

Puzzle 27: What is the output of this code?

In verbose programming languages such as Java, you

have to iterate over a list to remove subsequent elements.
In Python, a simple one-liner does that for you. Use the
slice notation to select a sequence of items in the list.
This is the lefthand side of your equation. Then overwrite
the selected sequence with the empty (or any other) list.
It is because of this kind of clarity and simplicity that
Python has become so popular nowadays.
5.27. CLEARING SUBLISTS 117

The correct solution 

[`a']

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 45 -10
1000 - 1500 27 -28
1500 - 2000 10 -45
>2000 8 -47

Your current Elo rating 

118 CHAPTER 5. FIFTY CODE PUZZLES

5.28 The Fibonacci Series

Puzzle 28
#############################
## id 346
## Puzzle Elo 1300
## Correctly solved 48 %
#############################

# Fibonacci series:
a, b = 0, 1
while b < 5:
print(b)
a, b = b, a + b

Puzzle 28: What is the output of this code?

The Fibonacci series was discovered by the Italian

mathematician Leonardo Fibonacci in 1202 and even ear-
lier by Indian mathematicians. The series appears in un-
expected areas such as economics, mathematics, art, and
nature.

In the puzzle, we give a simple algorithm to calculate

the Fibonacci numbers. The series starts with the Fi-
bonacci numbers zero and one. The algorithm calculates
the next element of the series as the sum of the previ-
ous two elements. For this, the algorithm only has to
5.28. THE FIBONACCI SERIES 119

keep track of the last two elements in the series. Thus,

we maintain two variables a and b, being the second last
and last element in the series, respectively. This com-
putation repeats until the while condition evaluates to
False, i.e., until b≥5.
For clarity of the code, I used the language feature
of iterable unpacking in the rst and the last line. This
feature works as follows. On the left-hand side of the
assignment, there is any sequence of variables. On the
right-hand side of the assignment, we specify the values
to be assigned to these variables.

Note that all expressions on the right-hand side are

rst evaluated before they are assigned. This is an impor-
tant property for our algorithm. Without this property,
the last line would be wrong as the expression a+b would
consider the wrong value for a.

The correct solution 

1 1 2 3
120 CHAPTER 5. FIFTY CODE PUZZLES

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 46 -9
1000 - 1500 30 -25
1500 - 2000 10 -45
>2000 8 -47

Your current Elo rating 

5.29. THE CONTINUE STATEMENT AND THE
MODULO OPERATOR 121

5.29 The continue Statement and

the Modulo Operator

Puzzle 29
#############################
## id 355
## Puzzle Elo 1311
## Correctly solved 54 %
#############################

for num in range(2, 8):

if not num % 2:
continue
print(num)

Puzzle 29: What is the output of this code?

This puzzle prints all odd values between two (in-

cluded) and eight (excluded). To achieve this, we check
whether the current value num can be divided by 2 with-
out remainder. Python, like other languages, uses the
percentage symbol % as modulo operator. This modulo
operator returns the remainder when dividing a number
n by another number x, i.e., n - (n//x) * x. For exam-
ple, it returns 12 % 2 = 0 if there is no remainder and
13 % 2 = 1 if the remainder is 1.
122 CHAPTER 5. FIFTY CODE PUZZLES

There is a second language feature in the puzzle, which

is the continue operator. This operator commands the
interpreter to terminate the current loop iteration. Then,
the interpreter proceeds with the next iteration.

Hence, if the loop variable has an even value, the in-

terpreter skips the print statement. If the loop variable
has an odd value, the interpreter skips the continue state-
ment.

The correct solution 

3
5
7

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 46 -9
1000 - 1500 31 -24
1500 - 2000 10 -45
>2000 8 -47

Your current Elo rating 

5.30. INDEXING REVISITED AND THE RANGE
SEQUENCE 123

5.30 Indexing Revisited and The

Range Sequence

Puzzle 30
#############################
## id 351
## Puzzle Elo 1346
## Correctly solved 52 %
#############################

print(range(5, 10)[-1])
print(range(0, 10, 3)[2])
print(range(-10, -100, -30)[1])

Puzzle 30: What is the output of this code?

If this book can teach you only one thing, it is a

thorough understanding of the most important Python
concepts such as indexing and slicing. I cannot empha-
size enough how important these concepts are for your
practical work. The goal of this puzzle is to strengthen
your understanding of these. The puzzle consists of three
quick tasks about indexing and the range function. Rep-
etition is an eective teacher!

The rst line prints the last element of the range se-
quence. A short reminder: the upper bound range pa-
124 CHAPTER 5. FIFTY CODE PUZZLES

rameter is not included in the sequence. The second line

prints the third element (not the second) of the range
sequence 0, 3, 6, 9. Thus, the step size is three as
dened in the last optional range parameter. The third
line prints the second element of the range sequence -10,
-40, -70 with step size -30.
Many nxters have problems with indexing or the
range function. One common mistake is that they select
the wrong element from the sequenceforgetting that
the rst element of any sequence has index 0, not index
1.

It is the procient use of the basics that dierentiates

excellent from average programmers.

The correct solution 

-40
5.30. INDEXING REVISITED AND THE RANGE
SEQUENCE 125

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 46 -9
1000 - 1500 33 -22
1500 - 2000 11 -44
>2000 8 -47

Your current Elo rating 

126 CHAPTER 5. FIFTY CODE PUZZLES
5.31. SEARCHING IN SORTED MATRIX 127

5.31 Searching in Sorted Matrix

Puzzle 31
#############################
## id 112
## Puzzle Elo 1353
## Correctly solved 41 %
#############################

def matrix_find(matrix, value):

if not matrix or not matrix[0]:
return False

j = len(matrix) - 1
for row in matrix:
while row[j] > value:
j = j - 1
if j == -1:
return False
if row[j] == value:
return True
return False

matrix = [[3, 4, 4, 6],

[6, 8, 11, 12],
[6, 8, 11, 15],
[9, 11, 12, 17]]
print(matrix_find(matrix=matrix, value=11))

Puzzle 31: What is the output of this code?

128 CHAPTER 5. FIFTY CODE PUZZLES

The puzzles are getting harder now. The challenge

is shifting from understanding syntactical to semantical
code snippets and algorithms. If you thoroughly mas-
ter these types of code puzzles, you will join the club
of advanced coders. Thus, you open up the opportunity
to work in one of the highest paid job industries in the
world.

This algorithm is a beautiful way to search a value in

a sorted matrix without visiting all values. In the next
paragraph, I describe the matrix concept and the sorted
property.

A matrix is a table of values consisting of rows and

columns. This puzzle represents it as a list of integer
lists. Hence, we can access matrix values with the index-
ing and slicing notation. Do you see the importance of
understanding the basics? The matrix is sorted as the
integers in the rows and columns increase monotonically
with the row and column number.

The function matrix_find takes a sorted integer ma-

trix and an integer value. It returns True if the matrix
contains the integer value. Otherwise, it returns False.

In the rst two lines, the algorithm checks whether

the matrix is empty and returns False if this is the case.
Then, the for loop iterates over rows of the matrix start-
ing with the rst row.

But instead of searching the whole matrix, the algo-

5.31. SEARCHING IN SORTED MATRIX 129

rithm uses a smarter strategy. It skips whole rows and

columns at a time using the sorted property.

The algorithm starts with the rst row and the last
column j = len(matrix) - 1. Then, it skips one col-
umn at-a-time by decreasing the parameter j monoton-
ically (j = j - 1). Why can it skip the whole column?
Because as long as the column value row[j] is larger
than the searched value value, all following elements of
column j are larger than the searched value (sorted prop-
erty). Thus, we are sure that our searched value is not
in column j and we can skip this column completely by
decreasing j.

If the column value row[j] is smaller than the searched

value, the algorithm skips this whole row by going to the
next row. Why can it skip the whole row? Because it
currently checks the largest value in the row. If this value
is smaller than the searched value, all other values are as
well.

In summary, the idea of this great algorithm from

2
Keith Schwartz is to skip either one row or one column
in each step. Thus, for a quadratic matrix with n rows
and columns, the algorithm inspects approximately 2n
2
cells. Note that a naive algorithm would inspect all n
cells which is much slower.

2 http://www.keithschwarz.com/interesting/code/
matrix-find/MatrixFind.python.html
130 CHAPTER 5. FIFTY CODE PUZZLES

The correct solution 

True

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 46 -9
1000 - 1500 33 -22
1500 - 2000 11 -44
>2000 8 -47

Your current Elo rating 

5.32. MAXIMUM PROFIT ALGORITHM 131

5.32 Maximum Prot Algorithm

Puzzle 32
#############################
## id 36
## Puzzle Elo 1407
## Correctly solved 65 %
#############################

def maximum_profit(prices):
'''Maximum profit of a single buying low and
,→ selling high'''

profit = 0
for i, buy_price in enumerate(prices):
sell_price = max(prices[i:])
profit = max(profit, sell_price -
,→ buy_price)
return profit

# Ethereum daily prices in Dec 2017 ($)

eth_prices = [455, 460, 465, 451, 414, 415, 441]
print(maximum_profit(prices=eth_prices))

Puzzle 32: What is the output of this code?

This puzzle presents an algorithmic problem with prac-

tical value for stock market analysis. Suppose you are
132 CHAPTER 5. FIFTY CODE PUZZLES

trading the cryptocurrency Ethereum. How much prot

can you make by buying low and selling high based on
historical data?

The function maximum_profit takes as input a se-

quence of prices, e.g., a week of Ethereum prices in De-
cember 2017. It returns the largest possible prot of
buying low and selling high.

The algorithm works as follows. It iterates over all

sequence values: each is a possible buying point (i.e.,
buy_price). Note that the enumerate function returns
both the index i of the next price in the sequence and
the price itself.

Next, the algorithm uses the index i of the current

buying point to get all potential selling points after buy-
ing. We use slicing to get these, i.e., prices[i:]. The
max function nds the highest selling point. For each
buying/selling pair (buy_price, sell_price), it calcu-
lates the prot as the dierence between the prices at the
sell_price-buy_price.
selling and the buying points, i.e.,
The variable profit maintains the largest possible prot:
$27 on $414 invested capital.

The correct solution 

27
5.32. MAXIMUM PROFIT ALGORITHM 133

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 36 -19
1500 - 2000 12 -43
>2000 8 -47

Your current Elo rating 

134 CHAPTER 5. FIFTY CODE PUZZLES

5.33 Bubble Sort Algorithm

Puzzle 33
#############################
## id 158
## Puzzle Elo 1458
## Correctly solved 67 %
#############################

def bubble_sort(lst):
'''Implementation of bubble sort
,→ algorithm'''

for border in range(len(lst)-1, 0, -1):

for i in range(border):
if lst[i] > lst[i + 1]:
lst[i], lst[i + 1] = lst[i + 1],
,→ lst[i]
return lst

list_to_sort = [27, 0, 71, 70, 27, 63, 90]

print(bubble_sort(lst=list_to_sort))

Puzzle 33: What is the output of this code?

The bubble sort algorithm works exactly as its name

suggests. It sorts an input list by treating each element
as a bubble that climbs up the list. Each bubble rises as
5.33. BUBBLE SORT ALGORITHM 135

long as it is greater than the list elements. If the bubble

element is smaller or equal than a list element x, the
bubble stops rising, and the larger list element x starts
to bubble up.

The precise algorithm works as follows. The outer

index variable border marks the index after which the
right-hand list elements are already sorted. The inner
index variable i goes from left to right until it reaches
the index variable border. On its way to the right, it
switches two subsequent list elements if the rst element
is larger than the second element. Hence, after the rst
pass, the largest element in the list is on the right. As
this right-most element is already sorted, we can reduce
the size of the list to be sorted by onei.e., decrement
the variable border. Next, the second largest element
will rise to the top and the procedure repeats.

Study this basic algorithm carefully. Every great coder

must know it.

The correct solution 

[0, 27, 27, 63, 70, 71, 90]

136 CHAPTER 5. FIFTY CODE PUZZLES

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 38 -17
1500 - 2000 14 -41
>2000 8 -47

Your current Elo rating 

5.34. JOINING STRINGS 137

5.34 Joining Strings

Puzzle 34
#############################
## id 367
## Puzzle Elo 1437
## Correctly solved 53 %
#############################

def concatenation(*args, sep="/"):

return sep.join(args)

print(concatenation("A", "B", "C", sep=","))

Puzzle 34: What is the output of this code?

String concatenation is the process of creating a string

by appending string arguments. The given function takes
an arbitrary number of string arguments as specied by
the *args keyword. The parameter sep declares the sep-
arator string to be used to glue together two strings. The
separator string comes as a keyword argument. The rea-
son is that the *args argument comprises an arbitrary
number of values. The keyword argument helps to dif-
ferentiate whether the last parameter is part of *args or
the sep argument.
138 CHAPTER 5. FIFTY CODE PUZZLES

The function concatenation is a wrapper for the

join function to concatenate strings. The join func-
tion is dened in the string object sep. It concatenates
an arbitrary number of strings using the separator to glue
them together. Both functions achieve the same thing,
but the rst may be more convenient because the sepa-
rator is a normal argument. Yet, you will nd yourself
using the join function on a regular basis without writ-
ing your own wrapper functions. So you may as well
learn its proper use now.

The correct solution 

A,B,C

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 37 -18
1500 - 2000 13 -42
>2000 8 -47

Your current Elo rating 

5.35. ARITHMETIC CALCULATIONS 139

5.35 Arithmetic Calculations

Puzzle 35
#############################
## id 320
## Puzzle Elo 1486
## Correctly solved 7 %
#############################

x = 5 * 3.8 - 1
print(x)

Puzzle 35: What is the output of this code?

This puzzle has only one challenge. But this challenge

is so hard that only 7% of all nxters can overcome it:
oating point operators.

Most nxters believe that the puzzle asks for the re-
sult of the computation here. But this is a trap! The
purpose of solving Python puzzles is to understand code
in a precise and deep manner. Deep understanding tells
you that the oat 3.80 causes the interpreter to per-
form oating point arithmetic. Thus, the result is not an
integeri.e., the value 18but a oati.e., the value
18.0.
These kinds of mistakes seem to be negligible but they
140 CHAPTER 5. FIFTY CODE PUZZLES

have important eects on the correctness of your code

base. So if you got this puzzle wrong, be grateful for the
lesson and go on.

The correct solution 

18.0

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 39 -16
1500 - 2000 15 -40
>2000 8 -47

Your current Elo rating 

5.36. BINARY SEARCH 141

5.36 Binary Search

Puzzle 36
#############################
id 159
## Puzzle Elo 1492
## Correctly solved 33 %
#############################

def bsearch(l, value):

lo, hi = 0, len(l)-1
while lo <= hi:
mid = (lo + hi) // 2
if l[mid] < value:
lo = mid + 1
elif value < l[mid]:
hi = mid - 1
else:
return mid
return -1

l = [0, 1, 2, 3, 4, 5, 6]
x = 6
print(bsearch(l,x))

Puzzle 36: What is the output of this code?

How to nd a value in a sorted list? The naive al-

142 CHAPTER 5. FIFTY CODE PUZZLES

gorithm compares each element in the list against the

searched value. For example, consider a list with 1024
elements. The naive algorithm performs 1024 compar-
isons in the worst case.

The function bsearch is a more eective way to nd a

n elements in the list, it needs to
value in a sorted list. For
perform only in the order of log(n) comparisons. Hence,
a list with 1024 elements would take Bsearch only up to
log(1024) = 10 comparisonsmaking it much faster

Why is Bsearch so fast? Bsearch uses the property

that the list is already sorted. It checks only the element
in the middle position between two indices lo and hi. If
this middle element is smaller than the searched value, all
left-hand elements will be smaller as well because of the
sorted list. The algorithm can skip all left-hand elements
by setting the lower index lo to the position right of the
middle element. If this middle element is larger than the
searched value, all right-hand elements will be larger as
well. Hence, we set the upper index hi to the position
left of the middle element. Only if the middle element
is exactly the same as the searched value, we return the
index of this position. This procedure is repeated until
we nd the searched value or there are no values left. In
each loop iteration, we reduce the search space, i.e., the
number of elements between lo and hi, by half.
5.36. BINARY SEARCH 143

The correct solution 

6

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 40 -15
1500 - 2000 15 -40
>2000 8 -47

Your current Elo rating 

144 CHAPTER 5. FIFTY CODE PUZZLES

5.37 Modifying Lists in Loops

Puzzle 37
#############################
## id 349
## Puzzle Elo 1504
## Correctly solved 58 %
#############################

words = ['cat', 'mouse', 'dog']

for word in words[:]:
if len(word) > 3:
words.insert(0, word)
print(words[0])

Puzzle 37: What is the output of this code?

How to modify a sequence while iterating over it? For

example, you want to prepare a data set of house prices
for a machine learning algorithm to predict the market
prices of new houses. Your goal is to remove the data
points with prices lower than $20, 000 to clean the data
of outliers.

This problem is not as simple as removing elements

from a sequence over which you iterate. Doing this can
lead to unspecied behavior as explained in the follow-
ing. Before entering the for loop, the Python interpreter
5.37. MODIFYING LISTS IN LOOPS 145

creates an iterator object. The iterator object provides

a method next() returning the next element in the se-
quence. To achieve this, the iterator extracts, at cre-
ation time, information like the size of the sequence. If
you modify the sequence on the go, this information
becomes invalid. For example, if the number of elements
changes at runtime, the iterator object may believe it is
ready, while there are still objects in it.

The puzzle presents one solution to this problem. The

code copies the list rst and iterates over the copy. With
this method, you can safely modify the original list as
this will not aect the copy in any way.

So how to copy the sequence? The most convenient

way to achieve this is by using the slice notation as shown
in the puzzle.

The correct solution 

mouse
146 CHAPTER 5. FIFTY CODE PUZZLES

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 40 -15
1500 - 2000 15 -40
>2000 8 -47

Your current Elo rating 

5.38. THE LAMBDA FUNCTION 147

5.38 The Lambda Function

Puzzle 38
#############################
## id 370
## Puzzle Elo 1558
## Correctly solved 89 %
#############################

def make_incrementor(n):
return lambda x: x + n

f = make_incrementor(42)
print(f(0))
print(f(1))

Puzzle 38: What is the output of this code?

This puzzle introduces an advanced language feature:

lambda functions. Lambda functions are rooted in the
mathematical area of lambda calculus. One of the pi-
oneers of this area was Alonzo Church. He introduced
lambda functions in 1936 even before the appearance of
the rst computers.

Lambda functions exist in a wide range of languages

for functional programming. They are not only at the
heart of functional programming languages, they are also
148 CHAPTER 5. FIFTY CODE PUZZLES

the basis of many advanced Python language features.

For example, the modern language Scala for parallel pro-
gramming combines traditional language elements (e.g.,
from Java) with functional elements (e.g., lambda func-
tions). So how do lambda functions work?

A lambda function is an anonymous function with-

out identier. After the lambda keyword, the function
takes one or more arbitrary arguments. The arguments
are comma-separated and nished by a colon. After the
colon follows a single expression. Yet, this expression
can consist of complex calculations using the specied
argument variables. The lambda function then returns
the result of this expression. Hence, lambda functions
are syntactical shortcuts for a subclass of normal Python
functions.

In the puzzle, the function make_incrementor cre-

ates a lambda function at runtime. The created lambda
function increases an element x by a xed value n. For ex-
ample, the incrementor function in the puzzle increments
a value by 42. We assign this function to the variable f.
Then we print the results when incrementing the values
0 and 1 by the incrementor 42.

The correct solution 

42
5.38. THE LAMBDA FUNCTION 149

43

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 42 -13
1500 - 2000 17 -38
>2000 8 -47

Your current Elo rating 

150 CHAPTER 5. FIFTY CODE PUZZLES

5.39 Multi-line Strings and the

New-line Character

Puzzle 39
#############################
## id 325
## Puzzle Elo 1623
## Correctly solved 71 %
#############################

print("""
A
B
C
""" == "\nA\nB\nC\n")

Puzzle 39: What is the output of this code?

What is going on in this puzzle? The basic idea is

to show two dierent ways of writing the same multi-line
string literal in Python.

The rst is the direct way to write a multi-line string

in Python: As a string with multiple code lines enclosed
by triple-quotes '''...''' or """...""".
The second is a more concise way to write the same
5.39. MULTI-LINE STRINGS AND THE NEW-LINE
CHARACTER 151

string. We specify the line breaks with the new line char-
acter '\n'.

These two ways of breaking lines in Python strings

are the basis for advanced features and code snippets.

The correct solution 

True

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 43 -12
1500 - 2000 20 -35
>2000 9 -46

Your current Elo rating 

152 CHAPTER 5. FIFTY CODE PUZZLES

5.40 Escaping

Puzzle 40
#############################
## id 323
## Puzzle Elo 1629
## Correctly solved 25 %
#############################

print('P"yt\'h"on')

Puzzle 40: What is the output of this code?

This puzzle introduces several Python language fea-

tures about quotes in string literals. It requires a clear
understanding of the concept of escaping. Escaping is
an important concept in most programming languages.
You are not an advanced coder without understanding
at least the basic idea of escaping.

Recap, strings can be enclosed either with single quotes

'...' or double quotes "...". These two options are
semantically equivalent, i.e., they do the same thing.

But what happens if you write, say, a small conver-

sation with direct speech?
5.40. ESCAPING 153

"Alice said: "Hey Bob!" and went on."

(wrong)

The double quotes cannot be a part of a string en-

closed in double quotes. Trying this ends the string pre-
maturely. Here, the best case is that the interpreter com-
plains about the strange syntax of the random character
sequence after the premature ending of your string.

Yet, there is an easy x. You can avoid this problem

by enclosing the string with single quotes:

'Alice said: "Hey Bob!" and went on.'

(right)

The double quotes can now be part of the string itself

without ending the string sequence. The opposite also
works, i.e., writing a single quote within a string enclosed
in double quotes.

So far so good. But there is still one question left that

is also the main reason why only 25% of nxters can solve
this puzzle: escaping. What if you want to put a single
quote within a string enclosed by single quotes?

In the puzzle, we solve this using the escape character:

the backslash \. When put before special characters like
the single quote, it escapes them. In other words, it
changes the meaning of these characters. For example,
154 CHAPTER 5. FIFTY CODE PUZZLES

the single quote has the meaning of starting or ending a

string. Only when escaped, the interpreter changes its
meaning to the normal single quote character.

The correct solution 

P"yt'h"on

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 43 -12
1500 - 2000 21 -34
>2000 9 -46

Your current Elo rating 

5.41. FIBONACCI ARITHMETIC 155

5.41 Fibonacci Arithmetic

Puzzle 41
#############################
## id 359
## Puzzle Elo 1661
## Correctly solved 60 %
#############################

def fibo(n):
"""Return list containing
Fibonacci series up to n.
"""

result = []
a, b = 0, 1
while a < n:
result.append(a)
a, b = b, a + b
return result

fib100 = fibo(100)
print(fib100[-1] ==
fib100[-2] + fib100[-3])

Puzzle 41: What is the output of this code?

Recap the Fibonacci series is the series of numbers

that arises when repeatedly summing up the last two
156 CHAPTER 5. FIFTY CODE PUZZLES

numbers starting from 0 and 1. The fibo function in the

puzzle calculates all Fibonacci numbers up to the func-
tion argument n. We use the concise method of iterable
unpacking to store the value of b in the variable a and
to calculate the new value of b as the sum of both. We
maintain the whole sequence in the list variable result
by appending the sequence value a to the end of the list.

The puzzle calculates the Fibonacci sequence up to

100 and stores the whole list in the variable fib100. But
to solve the puzzle, you do not have to calculate the whole
sequence. The print statement only compares whether
the last element is equal to the sum of the second and
third last element in the sequence. This is true by de-
nition of the Fibonacci series.

Humans can solve this puzzle easily using logic and

strategic thinking. The Python interpreter, however, must
take the brute-force approach of calculating everything
from scratch. This nicely demonstrates your role as a
computer programmer. You are the guiding hand with
unlimited power at your ngertips. But you must use
your power wisely because the computer will do exactly
what you ask it to do.

The correct solution 

True
5.41. FIBONACCI ARITHMETIC 157

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 44 -11
1500 - 2000 22 -33
>2000 9 -46

Your current Elo rating 

158 CHAPTER 5. FIFTY CODE PUZZLES

5.42 Quicksort

Puzzle 42
#############################
## id 195
## Puzzle Elo 1672
## Correctly solved 67 %
#############################

def qsort1(L):
if L:
return qsort1([x for x in L[1:] if x <
,→ L[0]]) + L[:1] \
+ qsort1([x for x in L[1:] if x
,→ >= L[0]])
return []

def qsort2(L):
if L:
return L[:1] + qsort2([x for x in L[1:]
,→ if x < L[0]]) \
+ qsort2([x for x in L[1:] if x
,→ >= L[0]])
return []

print(qsort1([0, 33, 22]))

print(qsort2([0, 33, 22]))

Puzzle 42: Which function correctly sorts the list?

5.42. QUICKSORT 159

This puzzle introduces a recursive algorithm to sort

lists. When executing the functions, you get the following
results.

print(qsort1([0,33,22])) > output: [0,

22, 33]
print(qsort2([0,33,22])) > output: [0,
33, 22]

So, based on this output, the function qsort1 cor-

rectly sorts the list. But why? The algorithm is a variant
of the popular quicksort algorithm. Quicksort selects a
pivot element from the list. In the puzzle, it selects the
rst element of the list, i.e., L[0]. Then, the algorithm
moves all elements that are smaller than the pivot to the
left side. Similarly, it moves elements that are larger or
equal than the pivot to the right side.

This is repeated in a recursive manner for the left and

the right lists. Suppose you create a new list as follows.
You put all elements that are smaller than the pivot on
the left, then the pivot, then all elements that are larger
or equal the pivot on the right. The resulting list feels a
bit more sorted, right? If the two sublists were already
sorted, the list would be perfectly sorted. This is where
the recursive call of qsort1 comes into play. It takes over
the problem of sorting each sublist by applying the same
scheme of pivoting and recursion to the sublist.
160 CHAPTER 5. FIFTY CODE PUZZLES

In contrast, the qsort2 function appends both sub-

lists to the right of the pivot element. Hence the list is
already unsorted after the rst recursion level.

Solving these kinds of puzzles regularly will boost

your code understanding skills. They not only train your
language understanding but also your conceptual think-
ing which is even more important for coders at any level.

The correct solution 

qsort1

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 44 -11
1500 - 2000 23 -32
>2000 9 -46

Your current Elo rating 

5.43. UNPACKING KEYWORD ARGUMENTS
WITH DICTIONARIES 161

5.43 Unpacking Keyword

Arguments with Dictionaries
Puzzle 43
#############################
## id 369
## Puzzle Elo 1673
## Correctly solved 30 %
#############################

def func(val1=3, val2=4, val3=6):

return val1 + val2 + val3

values = {"val1":9, "val3":-1}

print(func(**values))

Puzzle 43: What is the output of this code?

Programming is about using lower-level functionality

to create higher-level functionality. In general, any pro-
gramming language is a collection of functions that in
turn build upon functions provided by the operating sys-
tem. You must master the art of building your own code
with the help of existing functionality. Do not reinvent
the wheel!

Functions are generic code snippets that can be tai-

lored to your needs via keyword arguments. The puzzle
162 CHAPTER 5. FIFTY CODE PUZZLES

shows a function that calculates the sum of three keyword

arguments. The keyword arguments are initialized with a
default value in case they are not dened by the function
caller. The puzzle introduces two concepts: dictionaries
and unpacking keyword arguments.

1) Dictionaries are Python data structures, dened

via the bracket notation {}, that store key-value pairs.
Python dictionaries work like real-world dictionaries: the
keys are the words and the values are the explanations.
You access the explanation to a given word via the index
table. Similarly, in a Python dictionary, you access the
values using the method of indexing. The indices (or
keys) can be strings, integers, or any other immutable
data type.

2) An interesting twist in the puzzle is to deliver key-

word arguments via a dictionary using the **-operator.
The **-operator unpacks the key-value pairs in the dic-
tionary and matches those with the keyword arguments.
As the second keyword argument val2 is not declared in
the dictionary, it is initialized to its default value.

The correct solution 

12
5.43. UNPACKING KEYWORD ARGUMENTS
WITH DICTIONARIES 163

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 44 -11
1500 - 2000 23 -32
>2000 9 -46

Your current Elo rating 

164 CHAPTER 5. FIFTY CODE PUZZLES

5.44 Innity

Puzzle 44
#############################
id 356
## Puzzle Elo 1701
## Correctly solved 40 %
#############################

print("Answer")
while True:
pass
print("42")

Puzzle 44: What is the output of this code?

The question in this puzzle is whether the second

print statement will ever be executed. The body of the
while loop consists of the pass statement. This state-
ment tells the interpreter to do nothing. Although the
while loop does nothing, the interpreter is trapped for-
ever because the while condition is True. Thus, our pro-
gram wastes scarce CPU cycles until the user interrupts
the execution. Hence, no execution path will execute
the second print statement. It is interesting that 60% of
nxters get this puzzle wrong.
5.44. INFINITY 165

The correct solution 

Answer

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 45 -10
1500 - 2000 25 -30
>2000 9 -46

Your current Elo rating 

166 CHAPTER 5. FIFTY CODE PUZZLES
5.45. GRAPH TRAVERSAL 167

5.45 Graph Traversal

Puzzle 45
#############################
## id 274
## Puzzle Elo 1729
## Correctly solved 44 %
#############################

def has_path(graph, v_start, v_end, path_len=0):

'''Graph has path from v_start to v_end'''

# Traverse each vertex only once

if path_len >= len(graph):
return False

# Direct path from v_start to v_end?

if graph[v_start][v_end]:
return True

# Indirect path via neighbor v_nbor?

for v_nbor, edge in
,→ enumerate(graph[v_start]):
if edge: # between v_start and v_nbor
if has_path(graph, v_nbor, v_end,
,→ path_len + 1):
return True

return False

# The graph represented as adjancy matrix

G = [[1, 1, 0, 0, 0],
[0, 1, 0, 0, 0],
[0, 0, 1, 0, 0],
[0, 1, 1, 1, 0],
[1, 0, 0, 1, 1]]
print(has_path(graph=G, v_start=3, v_end=0))

Puzzle 45: Is there a path between vertices 3 and 0?

168 CHAPTER 5. FIFTY CODE PUZZLES

A simple and eective way to grow your computer

science skills is to master the basics. Knowing the basics
sets apart the great coders from the merely intermediate
ones. One such basic area in computer science is graph
theorywhich we address in this puzzle.

So rst things rst: what is a graph? You already

know data structures like lists, sets, and dictionaries.
These data structures are denoted as complex data struc-
tures not because they're dicult to understand but
because they build upon other data structures. A graph
is just another complex data structure for relational data.

Relational data consists of edges and vertices. Each

vertex stands in one or more relations with other ver-
tices. An example for relational data is the Facebook
social graph. Facebook represents users as vertices and
friendship relations as edges. Two users are connected
via an edge in the graph if they are (Facebook) friends.

How to maintain a graph data structure in the code?

The puzzle uses an adjacency matrix as graph data struc-
ture G. Each row i in the matrix stores the out-neighbors
of vertex i. And each column j stores the in-neighbors of
vertex j. Thus, there is an edge from vertex i to vertex
j, if G[i][j]==1.
How to determine whether there is a path between
two vertices? The function find_path(graph, v_start,
v_end, path_len) checks whether there is a direct or in-
5.45. GRAPH TRAVERSAL 169

direct path between two vertices v_start and v_end in

graph. We know that there is a direct path between
v_start and v_end if both are already neighbors, i.e.,
graph[v_start][v_end]==1.
However, even if there is not a direct path, there could
be an indirect path between vertices v_start and v_end.
To check this, the algorithm uses a recursive approach.
v_nbor
Specically, there is an indirect path if a vertex
exists such that there is a path v_start → v_nbor →
... → v_end.
The variable path_len stores the length of the cur-
rent path. We increment it in each recursion level as the
current path length increases by one. Note that all paths
with length ≥n consist of at least n vertices. In other
words, at least one vertex is visited twice and a cycle ex-
ists in this recursion instance. Hence, we skip recursion
for paths with length greater or equal than the number
of vertices in the graph.

This puzzle asks whether there is a path between 3

and 0. If you understand what the code is doing, it suf-
ces to look at the adjacency matrix G. There is a direct
path from vertex 3 to vertices 1 and 2 (and to itself ). But
neither vertex 1 nor 2 has any out-neighbors. Therefore,
there is no path from vertex 3 to any other vertex (be-
sides vertices 1 and 2).
170 CHAPTER 5. FIFTY CODE PUZZLES

The correct solution 

False

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 45 -10
1500 - 2000 26 -29
>2000 9 -46

Your current Elo rating 

5.46. LEXICOGRAPHICAL SORTING 171

5.46 Lexicographical Sorting

Puzzle 46
#############################
## id 371
## Puzzle Elo 1748
## Correctly solved 44 %
#############################

pairs = [(1, 'one'),

(2, 'two'),
(3, 'three'),
(4, 'four')]

# lexicographical sorting (ascending)

pairs.sort(key=lambda pair: pair[1])
print(pairs[0][1])

Puzzle 46: What is the output of this code?

The high Elo indicates that only experienced Python

coders can solve this puzzle. There are two barriers to
overcome.

First, the lambda function seems to be an abstract

concept. Yet, it is only old wine in a new bottle. A
lambda function is nothing but an anonymous function
with a special syntax. The variable name(s) between the
172 CHAPTER 5. FIFTY CODE PUZZLES

lambda keyword and the colon (:) dene the function ar-
guments. The body after the colon uses the arguments
to dene the return value of the function. In the puzzle,
we use the lambda function as a key for the sorting func-
tion. The key denes that the list should be sorted by
the second value of the tuple, which is a string.

Second, we are not sorting by ascending integers, i.e.,

1, 2, 3, 4, but by ascending strings according to their
position in the alphabet, i.e., 'four', 'one', 'three',
'two'. So the second tuple element from the rst list
element is 'four'.

The correct solution 

four

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 45 -10
1500 - 2000 27 -28
>2000 10 -45

Your current Elo rating 

5.46. LEXICOGRAPHICAL SORTING 173
174 CHAPTER 5. FIFTY CODE PUZZLES

5.47 Chaining of Set Operations

Puzzle 47
#############################
## id 399
## Puzzle Elo 1749
## Correctly solved 40 %
#############################

# popular instagram accounts

# (millions followers)
inst = {"@instagram":232,
"@selenagomez":133,
"@victoriassecret":59,
"@cristiano":120,
"@beyonce":111,
"@nike":76}

# popular twitter accounts

# (millions followers)
twit = {"@cristiano":69,
"@barackobama":100,
"@ladygaga":77,
"@selenagomez":56,
"@realdonaldtrump":48}

inst_names = set(filter(lambda key:

,→ inst[key]>60, inst.keys()))
twit_names = set(filter(lambda key:
,→ twit[key]>60, twit.keys()))

superstars = inst_names.intersection(twit_names)
print(list(superstars)[0])

Puzzle 47: What is the output of this code?

5.47. CHAINING OF SET OPERATIONS 175

You will use or have already used the concepts in-

troduced in this puzzle. They are elementary pieces of
knowledge for any Python programmer. There are three
basic concepts in the puzzle.

First, we have the two dictionaries mapping an ac-

count name to the number of followers. For example,
Cristiano Ronaldo (key: "@cristiano") has 120 million
Instagram followers. In contrast to lists, dictionaries al-
low fast data access. You can retrieve each item with only
one operation without having to iterate over the whole
data structure. In the words of a computer scientist: the
dictionary access has constant runtime complexity.

Second, the lter function returns a new sequence in

which each item matches a dened characteristic. The
lter function takes two arguments. The rst argument
is a function that returns a boolean valueTrue or False:
True if a sequence element should be included and False
otherwise. The second argument is the sequence to be
ltered.

Third, intersecting sets s1 and s2 returns a new set

that contains elements that are in both sets s1 and s2.
The only star that has more than 60 million Insta-
gram AND twitter followers is Cristiano Ronaldo.
176 CHAPTER 5. FIFTY CODE PUZZLES

The correct solution 

@cristiano

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 45 -10
1500 - 2000 27 -28
>2000 10 -45

Your current Elo rating 

5.48. BASIC SET OPERATIONS 177

5.48 Basic Set Operations

Puzzle 48
#############################
## id 390
## Puzzle Elo 1755
## Correctly solved 60 %
#############################

words_list = ["bitcoin",
"cryptocurrency",
"wallet"]
crawled_text = '''
Research produced by the University of
Cambridge estimates that in 2017,
there are 2.9 to 5.8 million unique
users using a cryptocurrency wallet,
most of them using bitcoin.
'''

split_text = crawled_text.split()
res1 = True in map(lambda word: word in
,→ split_text, words_list)
res2 = any(word in words_list for word in
,→ split_text)
print(res1 == res2)

Puzzle 48: What is the output of this code?

After executing the code in the puzzle, both res1 and

178 CHAPTER 5. FIFTY CODE PUZZLES

res2 store whether the variable crawled_text contains a

word from the word_list. I explain both ways to achieve
this in the following.

res1: The map function checks for each element word

word_list whether word is an element of the split
in the
crawled_text. The default split function divides the
string along the whitespaces. The result is an iterable
with three booleans, one for each word in the word_list.
Finally, we check whether one of them is True.

res2: The any function checks whether there is an

element in the iterable that is True. As soon as it nds
such a True value, this function returns True. Note that
it is more ecient to use the any function to do this
instead of performing a list iteration. After checking for
the rst word `bitcoin', the function already returns True.

The correct solution 

True
5.48. BASIC SET OPERATIONS 179

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 45 -10
1500 - 2000 28 -27
>2000 10 -45

Your current Elo rating 

180 CHAPTER 5. FIFTY CODE PUZZLES

5.49 Simple Unicode Encryption

Puzzle 49
#############################
## id 391
## Puzzle Elo 1763
## Correctly solved 66 %
#############################

def encrypt(text):
encrypted = map(lambda c: chr(ord(c) + 2),
,→ text)
return ''.join(encrypted)

def decrypt(text):
decrypted = map(lambda c: chr(ord(c) - 2),
,→ text)
return ''.join(decrypted)

s = "xtherussiansarecomingx"
print(decrypt(encrypt(encrypt(s))) ==
,→ encrypt(s))

Puzzle 49: What is the output of this code?

You already know that computers only operate on 0s

and 1s. Every single character in a string is encoded as
5.49. SIMPLE UNICODE ENCRYPTION 181

a sequence of 0s and 1s. Unicode is one such encoding

that maps a bunch of zeros and ones (a binary ordinal
value) to a symbol that you can read (a character). The
Unicode table assigns one binary or decimal value to each
character. For example, the Unicode value 41 encodes the
value 'A' and the Unicode value 42 the value 'B'.
With Unicode, we create our own secret language
via encryption and decryption functions. The functions
encrypt and decrypt operate on a string literal s1. To
encrypt or decrypt a string, we shift each character by
two Unicode positions. The encrypt function shifts the
string to the right, the decrypt function shifts it to the
left.

We use the map function to implement this shift for

each character in the string s1. Using the built-in func-
tion ord(), shifting a character is as simple as adding a
bias value to the Unicode value of the respective charac-
ter.

The result of both encryption and decryption is a se-

quence type. Hence, we join the sequence with the empty
string as a separator to receive the nal encrypted or de-
crypted string.

By calling the function encrypt() twice, the string

is simply shifted by 2 + 2 = 4 positions in the Unicode
table. Hence, the result of a double encryption plus a
single decryption is the same as a single decryption, i.e.,
182 CHAPTER 5. FIFTY CODE PUZZLES

2 + 2 − 2 = 2.

The correct solution 

True

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 46 -9
1500 - 2000 28 -27
>2000 10 -45

Your current Elo rating 

5.50. THE GUESS AND CHECK FRAMEWORK 183

5.50 The Guess and Check

Framework

Puzzle 50
#############################
## id 400
## Puzzle Elo 1780
## Correctly solved 56 %
#############################

import random

def guess(a, b):

return random.randint(a, b)

def check(x, y):

return y ** 2 == x

x = 100
left, right = 0, x
y = guess(left, right)
while not check(x, y):
y = guess(left, right)
print(y)

Puzzle 50: What is the output of this code?

184 CHAPTER 5. FIFTY CODE PUZZLES

The method of guess and check is a good starting

point for designing a new algorithm. The algorithm is
simple, parallelizable to thousands of cores, and well es-
tablished in theory. The runtime of the algorithm can
often be analyzed statistically.

The idea is to rst guess (generate) a possible solution

and then check whether it is correct (or acceptable). For
example, Bitcoin miners guess the solution to a complex
problem. If they nd a solution, a new bitcoin is created.
When generating the solution, it is common to use ran-
domization. The eciency of the algorithm depends on
how informed the guessing is. The better you guess, the
more ecient the algorithm becomes.

The puzzle nds an integer solution for the square

root of an input number x. The guess method generates
a random number y between 0 and x. It is not informed.
The check method checks whether the number y is the
square root.

The correct solution 

10
5.50. THE GUESS AND CHECK FRAMEWORK 185

Add this to your current Elo rating 

Your Elo Correct Incorrect
0 - 500 47 -8
500 - 1000 47 -8
1000 - 1500 46 -9
1500 - 2000 29 -26
>2000 10 -45

Your current Elo rating 

 6

Final Remarks

Congratulations, you made it through 50 code puzzles

and you have signicantly improved your skills in read-
ing and understanding code. By now, you should have a
fair estimate of your skill level in comparison to others
be sure to check out Table 3.1 again to get the respective
rank for your Elo rating. This book is all about push-
ing you from beginner to intermediate coding level. In
follow-up books, we address the advanced level with more
dicult puzzles.

Consistent eort and persistence is the key to success.

If you feel that solving code puzzles has advanced your
skills, make it a daily habit to solve a Python puzzle
and watch the related video that is given on the Finxter
web app. This habit alone will push your coding skills
through the roofand provide a comfortable living for

186
 187

you and your family in a highly protable profession.

Build this habit into your lifee.g., use your morning
coee break routineand you will soon become one of
the best programmers in your environment.

Where to go from here? I am publishing a fresh code

puzzle every couple of days on our website finxter.com.
All puzzles are available for free. My goal with Finxter
is to make learning to code more ecient, more individ-
ualized to your precise skill level, and more accessible
that's why I also post regular puzzles on our Facebook
page. For any feedback, question, or problem you strug-
gle and need help with, please send me an email to info@
finxter.com. If you want to grow your Python skills
on autopilot, register for our free puzzle newsletter at
app.finxter.com/accounts/register/.
Finally, I would like to express my deep gratitude
that you have spent your time solving code puzzles and
reading this book. Above everything else, I value your
time. The ultimate goal of any good textbook should be
to save, not take, your time. By working through this
textbook, you have gained insights about your coding
skill level and I hope that you have experienced a positive
return on invested time and money. Now, please keep
investing in yourself and stay active within the Finxter
community.